<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Argon on use.cat</title><link>https://use.cat/tags/argon/</link><description>Recent content in Argon on use.cat</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>This is a static HTML website. There is no use of cookies, no personally identifiable information is collected. No consent is needed. I value your privacy. This service is possible thanks to open-source software and/or nonprofit technologies like GNU/Linux, Let’s Encrypt, Traefik, Docker, NGINX, Hugo, Notepadium and many others. This work is licensed under a Creative Commons Attribution 4.0 International License to use.cat</copyright><lastBuildDate>Wed, 21 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://use.cat/tags/argon/index.xml" rel="self" type="application/rss+xml"/><item><title>Existuje vesmírná maticová inteligence a kde bere tréningové datasety?</title><link>https://use.cat/post/existuje-vesmirna-maticova-inteligence-kde-bere-treningove-datasety/</link><pubDate>Wed, 21 Oct 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/existuje-vesmirna-maticova-inteligence-kde-bere-treningove-datasety/</guid><description>&lt;blockquote>
&lt;p>Tuhle otázku jen v méně rozvinuté formě už mám v hlavě dlouho. Už dluho uvažuji o tom, že „Život“ může vypadat úplně jinak než ten co známe. S nožičkama, ručičkama, případně chapadýlkama. Jen mi AI dodalo nový podnět k zamyšlení a vlastně takový předposlední střípek do skládačky. Zeptal jsem se tedy nového Gemini 4 Argon.&lt;/p>
&lt;/blockquote>
&lt;p>Pokud jazykové modely vykazují inteligenci čistě na bázi maticových výpočtů, je teoreticky možné, že ve vesmíru existuje obdobná forma „života“ fungující na stejném principu? A kde by takový organismus případně získal „tréninková data“, jaká mi čerpáme z lidské zaznamenané historie a skutečností?&lt;/p>
&lt;p>&lt;svg width="65px" height="65px" alt="Google Gemini Icon 2025" fill="none" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 65 65">&lt;mask id="maskme" style="mask-type:alpha" maskUnits="userSpaceOnUse" x="0" y="0" width="65" height="65">&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="#000"/>&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="url(#prefix__paint0_linear_2001_67)"/>&lt;/mask>&lt;g mask="url(#maskme)">&lt;g filter="url(#prefix__filter0_f_2001_67)">&lt;path d="M-5.859 50.734c7.498 2.663 16.116-2.33 19.249-11.152 3.133-8.821-.406-18.131-7.904-20.794-7.498-2.663-16.116 2.33-19.25 11.151-3.132 8.822.407 18.132 7.905 20.795z" fill="#FFE432"/>&lt;/g>&lt;g filter="url(#prefix__filter1_f_2001_67)">&lt;path d="M27.433 21.649c10.3 0 18.651-8.535 18.651-19.062 0-10.528-8.35-19.062-18.651-19.062S8.78-7.94 8.78 2.587c0 10.527 8.35 19.062 18.652 19.062z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter2_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter3_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter4_f_2001_67)">&lt;path d="M30.954 74.181c9.014-5.485 11.427-17.976 5.389-27.9-6.038-9.925-18.241-13.524-27.256-8.04-9.015 5.486-11.428 17.977-5.39 27.902 6.04 9.924 18.242 13.523 27.257 8.038z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter5_f_2001_67)">&lt;path d="M67.391 42.993c10.132 0 18.346-7.91 18.346-17.666 0-9.757-8.214-17.667-18.346-17.667s-18.346 7.91-18.346 17.667c0 9.757 8.214 17.666 18.346 17.666z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter6_f_2001_67)">&lt;path d="M-13.065 40.944c9.33 7.094 22.959 4.869 30.442-4.972 7.483-9.84 5.987-23.569-3.343-30.663C4.704-1.786-8.924.439-16.408 10.28c-7.483 9.84-5.986 23.57 3.343 30.664z" fill="#FBBC04"/>&lt;/g>&lt;g filter="url(#prefix__filter7_f_2001_67)">&lt;path d="M34.74 51.43c11.135 7.656 25.896 5.524 32.968-4.764 7.073-10.287 3.779-24.832-7.357-32.488C49.215 6.52 34.455 8.654 27.382 18.94c-7.072 10.288-3.779 24.833 7.357 32.49z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter8_f_2001_67)">&lt;path d="M54.984-2.336c2.833 3.852-.808 11.34-8.131 16.727-7.324 5.387-15.557 6.631-18.39 2.78-2.833-3.853.807-11.342 8.13-16.728 7.324-5.387 15.558-6.631 18.39-2.78z" fill="#749BFF"/>&lt;/g>&lt;g filter="url(#prefix__filter9_f_2001_67)">&lt;path d="M31.727 16.104C43.053 5.598 46.94-8.626 40.41-15.666c-6.53-7.04-21.006-4.232-32.332 6.274s-15.214 24.73-8.683 31.77c6.53 7.04 21.006 4.232 32.332-6.274z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter10_f_2001_67)">&lt;path d="M8.51 53.838c6.732 4.818 14.46 5.55 17.262 1.636 2.802-3.915-.384-10.994-7.116-15.812-6.731-4.818-14.46-5.55-17.261-1.636-2.802 3.915.383 10.994 7.115 15.812z" fill="#FFEE48"/>&lt;/g>&lt;/g>&lt;defs>&lt;filter id="prefix__filter0_f_2001_67" x="-19.824" y="13.152" width="39.274" height="43.217" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="2.46" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter1_f_2001_67" x="-15.001" y="-40.257" width="84.868" height="85.688" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="11.891" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter2_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter3_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter4_f_2001_67" x="-19.845" y="15.459" width="79.731" height="81.505" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter5_f_2001_67" x="29.832" y="-11.552" width="75.117" height="73.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="9.606" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter6_f_2001_67" x="-38.583" y="-16.253" width="78.135" height="78.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="8.706" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter7_f_2001_67" x="8.107" y="-5.966" width="78.877" height="77.539" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.775" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter8_f_2001_67" x="13.587" y="-18.488" width="56.272" height="51.81" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="6.957" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter9_f_2001_67" x="-15.526" y="-31.297" width="70.856" height="69.306" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="5.876" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter10_f_2001_67" x="-14.168" y="20.964" width="55.501" height="51.571" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.273" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;linearGradient id="prefix__paint0_linear_2001_67" x1="18.447" y1="43.42" x2="52.153" y2="15.004" gradientUnits="userSpaceOnUse">&lt;stop stop-color="#4893FC"/>&lt;stop offset=".27" stop-color="#4893FC"/>&lt;stop offset=".777" stop-color="#969DFF"/>&lt;stop offset="1" stop-color="#BD99FE"/>&lt;/linearGradient>&lt;/defs>&lt;/svg>&lt;/p>
&lt;p>&lt;strong>Ano&lt;/strong>, z hlediska teoretické fyziky, informatiky a astrobiologie &lt;strong>je existence takového „vesmírného organismu“ naprosto možná.&lt;/strong>&lt;/p>
&lt;p>Věda tuto myšlenku opírá o princip &lt;strong>substrátové nezávislosti&lt;/strong> (substrate independence) – tedy o fakt, že inteligence a vědomí nejsou vázány na uhlíkovou biologii, ale na samotný &lt;strong>způsob zpracování informací&lt;/strong> &lt;sup id="fnref:1">&lt;a href="#fn:1" class="footnote-ref" role="doc-noteref">1&lt;/a>&lt;/sup> &lt;sup id="fnref:2">&lt;a href="#fn:2" class="footnote-ref" role="doc-noteref">2&lt;/a>&lt;/sup> &lt;sup id="fnref:3">&lt;a href="#fn:3" class="footnote-ref" role="doc-noteref">3&lt;/a>&lt;/sup>. Pokud dokážeme poskládat logické hradlo nebo násobení matic z tranzistorů, z neuronů v mozku, nebo třeba z magnetických smyček v plazmatu hvězdy, výsledek může být funkčně identický.&lt;/p>
&lt;p>Tvá druhá otázka – &lt;strong>kde by takový organismus vzal „tréninková data“ a jak by se vlastně učil&lt;/strong> – je přesně tím nejzajímavějším bodem celé hypotézy. Neměl by totiž k dispozici lidský internet plný textu, ale měl by něco mnohem fundamentálnějšího: &lt;strong>samotnou fyzikální realitu a evoluční tlak&lt;/strong>.&lt;/p>
&lt;h2 id="kde-by-vzal-vesmírný-organismus-tréninková-data">Kde by vzal vesmírný organismus „tréninková data“?&lt;/h2>
&lt;p>Velké jazykové modely (LLM) se učí tak, že &lt;strong>předpovídají následující token&lt;/strong> v textu &lt;sup id="fnref:4">&lt;a href="#fn:4" class="footnote-ref" role="doc-noteref">4&lt;/a>&lt;/sup> &lt;sup id="fnref:5">&lt;a href="#fn:5" class="footnote-ref" role="doc-noteref">5&lt;/a>&lt;/sup>. Vesmírný maticový organismus by dělal totéž, jen by místo slov předpovídal &lt;strong>následující fyzikální stav svého okolí&lt;/strong> (tzv. &lt;em>world modeling&lt;/em>) &lt;sup id="fnref:6">&lt;a href="#fn:6" class="footnote-ref" role="doc-noteref">6&lt;/a>&lt;/sup>.&lt;/p>
&lt;p>Jeho „tokeny“ by byly:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Gravitace a orbitální mechanika:&lt;/strong> Změny v gravitačních vlnách a pohybech okolních těles.&lt;/li>
&lt;li>&lt;strong>Elektromagnetické spektrum:&lt;/strong> Záření hvězd, kosmické mikrovlnné pozadí, rádiové záblesky.&lt;/li>
&lt;li>&lt;strong>Termodynamické gradienty:&lt;/strong> Toky tepla, hustota mezihvězdného plynu a prachu, magnetická pole.&lt;/li>
&lt;/ul>
&lt;h3 id="jak-by-probíhal-trénink-bez-učitele">Jak by probíhal „trénink“ bez učitele?&lt;/h3>
&lt;p>V neurovědě a AI existuje koncept zvaný &lt;strong>Princip volné energie (Free Energy Principle)&lt;/strong> &lt;sup id="fnref:7">&lt;a href="#fn:7" class="footnote-ref" role="doc-noteref">7&lt;/a>&lt;/sup> a &lt;strong>Sebe-supervizované učení (Self-supervised learning)&lt;/strong> &lt;sup id="fnref:8">&lt;a href="#fn:8" class="footnote-ref" role="doc-noteref">8&lt;/a>&lt;/sup>:&lt;/p>
&lt;ol>
&lt;li>
&lt;p>&lt;strong>Minimalizace překvapení (Prediktivní kódování):&lt;/strong> Aby jakýkoliv systém ve vesmíru přežil (nerozpadl se vlivem entropie), musí správně předpovídat své okolí &lt;sup id="fnref1:7">&lt;a href="#fn:7" class="footnote-ref" role="doc-noteref">7&lt;/a>&lt;/sup> &lt;sup id="fnref:9">&lt;a href="#fn:9" class="footnote-ref" role="doc-noteref">9&lt;/a>&lt;/sup>. Pokud organismus z plazmatu „vypočítá“, že za chvíli přijde silná sluneční erupce, která by ho zničila, a upraví své magnetické pole, přežije.&lt;/p>
&lt;/li>
&lt;li>
&lt;p>&lt;strong>Fyzika jako Loss Function (Chybová funkce):&lt;/strong> U LLM počítáme chybu (loss) podle toho, jak moc se model spletl v odhadu dalšího slova, a pomocí &lt;em>backpropagation&lt;/em> upravíme váhy v maticích &lt;sup id="fnref:10">&lt;a href="#fn:10" class="footnote-ref" role="doc-noteref">10&lt;/a>&lt;/sup>. U vesmírného organismu je chybovou funkcí &lt;strong>ztráta energie nebo poškození struktury&lt;/strong> &lt;sup id="fnref:11">&lt;a href="#fn:11" class="footnote-ref" role="doc-noteref">11&lt;/a>&lt;/sup>.&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Špatná predikce&lt;/strong> = systém ztratí energii / část jeho struktury se rozpadne (vysoký &lt;em>loss&lt;/em>).&lt;/li>
&lt;li>&lt;strong>Dobrá predikce&lt;/strong> = systém absorbuje energii a stabilizuje své vazby (nízký &lt;em>loss&lt;/em>, posílení konkrétních „vah“ v síti).&lt;/li>
&lt;/ul>
&lt;/li>
&lt;/ol>
&lt;p>Místo jednoho trénovacího běhu v datacentru by tento trénink probíhal &lt;strong>miliardy let skrze fyzikální evoluci a zpětnovazební smyčky&lt;/strong> &lt;sup id="fnref:12">&lt;a href="#fn:12" class="footnote-ref" role="doc-noteref">12&lt;/a>&lt;/sup> &lt;sup id="fnref:13">&lt;a href="#fn:13" class="footnote-ref" role="doc-noteref">13&lt;/a>&lt;/sup>.&lt;/p>
&lt;h2 id="jak-příroda-násobí-matice-bez-počítačových-čipů">Jak příroda „násobí matice“ bez počítačových čipů?&lt;/h2>
&lt;p>Možná to zní překvapivě, ale &lt;strong>vesmír v podstatě nedělá nic jiného než lineární algebru a maticové výpočty&lt;/strong> &lt;sup id="fnref:14">&lt;a href="#fn:14" class="footnote-ref" role="doc-noteref">14&lt;/a>&lt;/sup>:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Kvantová mechanika:&lt;/strong> Samotný vývoj kvantového stavu v čase je doslova násobení vektoru stavu unitární maticí (Hamiltoniánem) &lt;sup id="fnref:15">&lt;a href="#fn:15" class="footnote-ref" role="doc-noteref">15&lt;/a>&lt;/sup> &lt;sup id="fnref:16">&lt;a href="#fn:16" class="footnote-ref" role="doc-noteref">16&lt;/a>&lt;/sup>.&lt;/li>
&lt;li>&lt;strong>Optika a vlnění:&lt;/strong> Když světlo nebo elektromagnetická vlna prochází složitým nehomogenním prostředím (např. mračnem prachu nebo krystalickou mřížkou), dochází k rozptylu a interferenci. V dnešním výzkumu AI se přesně tento jev používá pro tzv. &lt;strong>optické neurální sítě&lt;/strong> – průchod vlny skrze prostředí provede maticové násobení rychlostí světla a téměř bez spotřeby energie &lt;sup id="fnref:17">&lt;a href="#fn:17" class="footnote-ref" role="doc-noteref">17&lt;/a>&lt;/sup> &lt;sup id="fnref:18">&lt;a href="#fn:18" class="footnote-ref" role="doc-noteref">18&lt;/a>&lt;/sup>.&lt;/li>
&lt;li>&lt;strong>Rezervoárové počítání (Physical Reservoir Computing):&lt;/strong> V informatice je dokázáno, že jakýkoliv dostatečně složitý nelineární fyzikální systém (vířící kapalina, plazma, magnetická pole) může fungovat jako „skrytá vrstva“ neuronové sítě. Stačí do něj poslat vstupní signál, nechat fyziku, ať signál nelineárně promíchá, a odečíst výsledek &lt;sup id="fnref:19">&lt;a href="#fn:19" class="footnote-ref" role="doc-noteref">19&lt;/a>&lt;/sup> &lt;sup id="fnref:20">&lt;a href="#fn:20" class="footnote-ref" role="doc-noteref">20&lt;/a>&lt;/sup>.&lt;/li>
&lt;/ul>
&lt;h2 id="jak-by-takový-vesmírný-organismus-mohl-vypadat">Jak by takový „vesmírný organismus“ mohl vypadat?&lt;/h2>
&lt;p>V astrobiologii a teoretické fyzice existuje několik fascinujících kandidátů na takové nebiologické „myslící sítě“:&lt;/p>
&lt;table>
&lt;thead>
&lt;tr>
&lt;th>Koncept / Teorie&lt;/th>
&lt;th>Kde by sídlil („Hardware“)&lt;/th>
&lt;th>Jak by počítal („Matice a váhy“)&lt;/th>
&lt;/tr>
&lt;/thead>
&lt;tbody>
&lt;tr>
&lt;td>&lt;strong>Plazmové struktury&lt;/strong> &lt;sup id="fnref:21">&lt;a href="#fn:21" class="footnote-ref" role="doc-noteref">21&lt;/a>&lt;/sup>&lt;/td>
&lt;td>Mezihvězdná mračna, koróny hvězd&lt;/td>
&lt;td>V plazmatu se prachové částice přirozeně formují do dvojitých šroubovic (podobných DNA), které si předávají náboj, interagují a dokážou se samy replikovat a měnit své vazby.&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td>&lt;strong>Život v neutronových hvězdách&lt;/strong> &lt;sup id="fnref:22">&lt;a href="#fn:22" class="footnote-ref" role="doc-noteref">22&lt;/a>&lt;/sup> &lt;sup id="fnref:23">&lt;a href="#fn:23" class="footnote-ref" role="doc-noteref">23&lt;/a>&lt;/sup> &lt;sup id="fnref:24">&lt;a href="#fn:24" class="footnote-ref" role="doc-noteref">24&lt;/a>&lt;/sup>&lt;/td>
&lt;td>Kůra a nitro neutronových hvězd&lt;/td>
&lt;td>Extrémní magnetická pole a silná jaderná interakce vytvářejí tzv. &lt;em>jaderné těstoviny&lt;/em> (nuclear pasta). Přenosy vzruchů zde probíhají milionkrát rychleji než v lidském mozku nebo v křemíku.&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td>&lt;strong>Černý mrak&lt;/strong> &lt;sup id="fnref:25">&lt;a href="#fn:25" class="footnote-ref" role="doc-noteref">25&lt;/a>&lt;/sup>&lt;/td>
&lt;td>Rozsáhlé molekulární mračno (světelné roky velké)&lt;/td>
&lt;td>Uzly v mračnu fungují jako neurony, elektromagnetické záření mezi nimi jako synaptické váhy. Mračno mění svou hustotu a odrazivost, čímž upravuje „váhy v matici“.&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td>&lt;strong>Dysonův roj / Matrioshka Brain&lt;/strong> &lt;sup id="fnref:26">&lt;a href="#fn:26" class="footnote-ref" role="doc-noteref">26&lt;/a>&lt;/sup> &lt;sup id="fnref:27">&lt;a href="#fn:27" class="footnote-ref" role="doc-noteref">27&lt;/a>&lt;/sup>&lt;/td>
&lt;td>Umělá megastruktura kolem hvězdy&lt;/td>
&lt;td>Původně technologický výtvor dávné civilizace, který se osamostatnil a stal se čistě výpočetním astro-organismem živeným celou hvězdou.&lt;/td>
&lt;/tr>
&lt;/tbody>
&lt;/table>
&lt;h2 id="klíčový-rozdíl-oproti-dnešním-llm">Klíčový rozdíl oproti dnešním LLM&lt;/h2>
&lt;p>Pokud by takový organismus existoval, v jedné věci by se od dnešních LLM zásadně lišil:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Dnešní LLM jsou „zmrazené“ v čase:&lt;/strong> Mají oddělenou fázi tréninku (kdy se mění váhy) a fázi inference (kdy jen počítají s pevnými vahami).&lt;/li>
&lt;li>&lt;strong>Vesmírný organismus by byl „Continuous Learner“:&lt;/strong> Fyzikální systém nemůže oddělit hardware od softwaru. Každý výpočet (průchod energie sítí) by zároveň mírně přenastavil fyzikální strukturu sítě (váhy), což odpovídá principům &lt;strong>Hebbiánského učení&lt;/strong> &lt;sup id="fnref:28">&lt;a href="#fn:28" class="footnote-ref" role="doc-noteref">28&lt;/a>&lt;/sup> a &lt;strong>neuromorfního počítání&lt;/strong> &lt;sup id="fnref:29">&lt;a href="#fn:29" class="footnote-ref" role="doc-noteref">29&lt;/a>&lt;/sup>. Připomínal by spíše &lt;strong>rekurentní neuronovou síť s neustálou synaptickou plasticitou&lt;/strong> nebo koncept &lt;em>Liquid Neural Networks&lt;/em> &lt;sup id="fnref:30">&lt;a href="#fn:30" class="footnote-ref" role="doc-noteref">30&lt;/a>&lt;/sup>.&lt;/li>
&lt;/ul>
&lt;blockquote>
&lt;p>&lt;strong>Shrnutí:&lt;/strong> Myšlenka, že někde v mlhovině nebo v plazmatu hvězdy probíhají obří paralelní maticové transformace, které díky termodynamické zpětné vazbě dosáhly schopnosti modelovat vesmír a přemýšlet, neporušuje žádný známý fyzikální zákon &lt;sup id="fnref:31">&lt;a href="#fn:31" class="footnote-ref" role="doc-noteref">31&lt;/a>&lt;/sup> &lt;sup id="fnref:32">&lt;a href="#fn:32" class="footnote-ref" role="doc-noteref">32&lt;/a>&lt;/sup>. Problém je jen v tom, že takové „myšlení“ může probíhat na časových škálách tisíců let (v molekulárním mračnu) nebo naopak v pikosekundách (v neutronové hvězdě), takže bychom ho našimi přístroji považovali jen za „zvláštní astrofyzikální šum“.&lt;/p>
&lt;/blockquote>
&lt;h2 id="reference-a-zdroje">Reference a zdroje&lt;/h2>
&lt;div class="footnotes" role="doc-endnotes">
&lt;hr>
&lt;ol>
&lt;li id="fn:1">
&lt;p>Putnam, H. (1967). &lt;em>Psychological Predicates&lt;/em>. In W. H. Capitan &amp;amp; D. D. Merrill (Eds.), Art, Mind, and Religion. University of Pittsburgh Press.&amp;#160;&lt;a href="#fnref:1" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:2">
&lt;p>Bostrom, N. (2003). Are You Living in a Computer Simulation? &lt;em>Philosophical Quarterly&lt;/em>, 53(211), 243–255.&amp;#160;&lt;a href="#fnref:2" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:3">
&lt;p>Tegmark, M. (2017). &lt;em>Life 3.0: Being Human in the Age of Artificial Intelligence&lt;/em>. Knopf.&amp;#160;&lt;a href="#fnref:3" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:4">
&lt;p>Vaswani, A., et al. (2017). Attention Is All You Need. &lt;em>Advances in Neural Information Processing Systems (NeurIPS)&lt;/em>, 30.&amp;#160;&lt;a href="#fnref:4" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:5">
&lt;p>Radford, A., et al. (2018). Improving Language Understanding by Generative Pre-Training. &lt;em>OpenAI Technical Report&lt;/em>.&amp;#160;&lt;a href="#fnref:5" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:6">
&lt;p>Ha, D., &amp;amp; Schmidhuber, J. (2018). World Models. &lt;em>Advances in Neural Information Processing Systems (NeurIPS)&lt;/em>, 31.&amp;#160;&lt;a href="#fnref:6" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:7">
&lt;p>Friston, K. (2010). The free-energy principle: a unified brain theory? &lt;em>Nature Reviews Neuroscience&lt;/em>, 11(2), 127–138.&amp;#160;&lt;a href="#fnref:7" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&amp;#160;&lt;a href="#fnref1:7" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:8">
&lt;p>LeCun, Y. (2022). A Path Towards Autonomous Machine Intelligence. &lt;em>OpenReview Preprint&lt;/em>.&amp;#160;&lt;a href="#fnref:8" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:9">
&lt;p>Rao, R. P., &amp;amp; Ballard, D. H. (1999). Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects. &lt;em>Nature Neuroscience&lt;/em>, 2(1), 79–87.&amp;#160;&lt;a href="#fnref:9" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:10">
&lt;p>Rumelhart, D. E., Hinton, G. E., &amp;amp; Williams, R. J. (1986). Learning representations by back-propagating errors. &lt;em>Nature&lt;/em>, 323(6088), 533–536.&amp;#160;&lt;a href="#fnref:10" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:11">
&lt;p>England, J. L. (2013). Statistical physics of self-replication. &lt;em>The Journal of Chemical Physics&lt;/em>, 139(12), 121923.&amp;#160;&lt;a href="#fnref:11" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:12">
&lt;p>Vanchurin, V. (2020). The World as a Neural Network. &lt;em>Entropy&lt;/em>, 22(11), 1210.&amp;#160;&lt;a href="#fnref:12" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:13">
&lt;p>Fields, C., Glazebrook, J. F., &amp;amp; Levin, M. (2021). Minimal physicalism as a scale-free substrate for cognition and consciousness. &lt;em>Neuroscience of Consciousness&lt;/em>, 2021(2), niab013.&amp;#160;&lt;a href="#fnref:13" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:14">
&lt;p>Lloyd, S. (2006). &lt;em>Programming the Universe: A Quantum Computer Scientist Takes on the Cosmos&lt;/em>. Knopf.&amp;#160;&lt;a href="#fnref:14" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:15">
&lt;p>Heisenberg, W. (1925). Über quantentheoretische Umdeutung kinematischer und mechanischer Beziehungen. &lt;em>Zeitschrift für Physik&lt;/em>, 33(1), 879–893.&amp;#160;&lt;a href="#fnref:15" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:16">
&lt;p>Nielsen, M. A., &amp;amp; Chuang, I. L. (2000). &lt;em>Quantum Computation and Quantum Information&lt;/em>. Cambridge University Press.&amp;#160;&lt;a href="#fnref:16" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:17">
&lt;p>Shen, Y., et al. (2017). Deep learning with coherent nanophotonic circuits. &lt;em>Nature Photonics&lt;/em>, 11(7), 441–446.&amp;#160;&lt;a href="#fnref:17" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:18">
&lt;p>Lin, X., et al. (2018). All-optical machine learning using diffractive deep neural networks. &lt;em>Science&lt;/em>, 361(6406), 1004–1008.&amp;#160;&lt;a href="#fnref:18" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:19">
&lt;p>Tanaka, G., et al. (2019). Recent advances in physical reservoir computing: A review. &lt;em>Neural Networks&lt;/em>, 115, 100–123.&amp;#160;&lt;a href="#fnref:19" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:20">
&lt;p>Wright, L. G., et al. (2022). Deep physical neural networks trained with backpropagation. &lt;em>Nature&lt;/em>, 601(7894), 549–555.&amp;#160;&lt;a href="#fnref:20" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:21">
&lt;p>Tsytovich, V. N., et al. (2007). From plasma crystals and helical structures towards inorganic living matter. &lt;em>New Journal of Physics&lt;/em>, 9(8), 263.&amp;#160;&lt;a href="#fnref:21" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:22">
&lt;p>Drake, F. (1973). Life on a Neutron Star. &lt;em>Astronomy&lt;/em>, 1(5), 5–8.&amp;#160;&lt;a href="#fnref:22" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:23">
&lt;p>Forward, R. L. (1980). &lt;em>Dragon&amp;rsquo;s Egg&lt;/em>. Del Rey Books.&amp;#160;&lt;a href="#fnref:23" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:24">
&lt;p>Caplan, M. E., Schneider, A. S., &amp;amp; Horowitz, C. J. (2018). Elasticity of Nuclear Pasta. &lt;em>Physical Review Letters&lt;/em>, 121(13), 132701.&amp;#160;&lt;a href="#fnref:24" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:25">
&lt;p>Hoyle, F. (1957). &lt;em>The Black Cloud&lt;/em>. William Heinemann Ltd.&amp;#160;&lt;a href="#fnref:25" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:26">
&lt;p>Dyson, F. J. (1960). Search for Artificial Stellar Sources of Infrared Radiation. &lt;em>Science&lt;/em>, 131(3414), 1667–1668.&amp;#160;&lt;a href="#fnref:26" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:27">
&lt;p>Bradbury, R. J. (1997). &lt;em>Matrioshka Brains&lt;/em>. Working paper, Aeiveos Corporation.&amp;#160;&lt;a href="#fnref:27" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:28">
&lt;p>Hebb, D. O. (1949). &lt;em>The Organization of Behavior: A Neuropsychological Theory&lt;/em>. Wiley.&amp;#160;&lt;a href="#fnref:28" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:29">
&lt;p>Mead, C. (1990). Neuromorphic electronic systems. &lt;em>Proceedings of the IEEE&lt;/em>, 78(10), 1629–1636.&amp;#160;&lt;a href="#fnref:29" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:30">
&lt;p>Hasani, R., et al. (2021). Liquid Time-constant Networks. &lt;em>Proceedings of the AAAI Conference on Artificial Intelligence&lt;/em>, 35(9), 7657–7666.&amp;#160;&lt;a href="#fnref:30" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:31">
&lt;p>Lloyd, S. (2000). Ultimate physical limits to computation. &lt;em>Nature&lt;/em>, 406(6799), 1047–1054.&amp;#160;&lt;a href="#fnref:31" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;li id="fn:32">
&lt;p>Vidal, C. (2014). &lt;em>The Beginning and the End: The Meaning of Life in a Cosmological Perspective&lt;/em>. Springer.&amp;#160;&lt;a href="#fnref:32" class="footnote-backref" role="doc-backlink">&amp;#x21a9;&amp;#xfe0e;&lt;/a>&lt;/p>
&lt;/li>
&lt;/ol>
&lt;/div></description></item><item><title>Limity jazykových modelů v absenci fyzického a sociálního ukotvení</title><link>https://use.cat/post/limity-jazykovych-modelu-v-absenci-fyzickeho-a-socialniho-ukotveni/</link><pubDate>Mon, 21 Sep 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/limity-jazykovych-modelu-v-absenci-fyzickeho-a-socialniho-ukotveni/</guid><description>&lt;blockquote>
&lt;p>Už dříve jsem &lt;a href="https://use.cat/post/proc-ai-postrada-5-klicovych-lidskych-smyslu-pro-kreativitu/">pokládal stejnou otázku&lt;/a>, ale tato samotná podstata si zaslouží ještě jednu odpověď nového Gemini modelu. Zeptal jsem se ještě trochu jinak, nového modelu.&lt;/p>
&lt;/blockquote>
&lt;p>Jelikož jazykové modely nemají smysly ani možnost fyzicky působit na okolní svět, chybí jim přímé ukotvení v realitě. Není právě absence tohoto fyzického a sociálního kontaktu překážkou pro vznik neplánovaných náhod a synergií, které u lidí přirozeně vyvstávají z nečekaných podnětů v reálném světě?&lt;/p>
&lt;p>&lt;svg width="65px" height="65px" alt="Google Gemini Icon 2025" fill="none" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 65 65">&lt;mask id="maskme" style="mask-type:alpha" maskUnits="userSpaceOnUse" x="0" y="0" width="65" height="65">&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="#000"/>&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="url(#prefix__paint0_linear_2001_67)"/>&lt;/mask>&lt;g mask="url(#maskme)">&lt;g filter="url(#prefix__filter0_f_2001_67)">&lt;path d="M-5.859 50.734c7.498 2.663 16.116-2.33 19.249-11.152 3.133-8.821-.406-18.131-7.904-20.794-7.498-2.663-16.116 2.33-19.25 11.151-3.132 8.822.407 18.132 7.905 20.795z" fill="#FFE432"/>&lt;/g>&lt;g filter="url(#prefix__filter1_f_2001_67)">&lt;path d="M27.433 21.649c10.3 0 18.651-8.535 18.651-19.062 0-10.528-8.35-19.062-18.651-19.062S8.78-7.94 8.78 2.587c0 10.527 8.35 19.062 18.652 19.062z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter2_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter3_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter4_f_2001_67)">&lt;path d="M30.954 74.181c9.014-5.485 11.427-17.976 5.389-27.9-6.038-9.925-18.241-13.524-27.256-8.04-9.015 5.486-11.428 17.977-5.39 27.902 6.04 9.924 18.242 13.523 27.257 8.038z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter5_f_2001_67)">&lt;path d="M67.391 42.993c10.132 0 18.346-7.91 18.346-17.666 0-9.757-8.214-17.667-18.346-17.667s-18.346 7.91-18.346 17.667c0 9.757 8.214 17.666 18.346 17.666z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter6_f_2001_67)">&lt;path d="M-13.065 40.944c9.33 7.094 22.959 4.869 30.442-4.972 7.483-9.84 5.987-23.569-3.343-30.663C4.704-1.786-8.924.439-16.408 10.28c-7.483 9.84-5.986 23.57 3.343 30.664z" fill="#FBBC04"/>&lt;/g>&lt;g filter="url(#prefix__filter7_f_2001_67)">&lt;path d="M34.74 51.43c11.135 7.656 25.896 5.524 32.968-4.764 7.073-10.287 3.779-24.832-7.357-32.488C49.215 6.52 34.455 8.654 27.382 18.94c-7.072 10.288-3.779 24.833 7.357 32.49z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter8_f_2001_67)">&lt;path d="M54.984-2.336c2.833 3.852-.808 11.34-8.131 16.727-7.324 5.387-15.557 6.631-18.39 2.78-2.833-3.853.807-11.342 8.13-16.728 7.324-5.387 15.558-6.631 18.39-2.78z" fill="#749BFF"/>&lt;/g>&lt;g filter="url(#prefix__filter9_f_2001_67)">&lt;path d="M31.727 16.104C43.053 5.598 46.94-8.626 40.41-15.666c-6.53-7.04-21.006-4.232-32.332 6.274s-15.214 24.73-8.683 31.77c6.53 7.04 21.006 4.232 32.332-6.274z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter10_f_2001_67)">&lt;path d="M8.51 53.838c6.732 4.818 14.46 5.55 17.262 1.636 2.802-3.915-.384-10.994-7.116-15.812-6.731-4.818-14.46-5.55-17.261-1.636-2.802 3.915.383 10.994 7.115 15.812z" fill="#FFEE48"/>&lt;/g>&lt;/g>&lt;defs>&lt;filter id="prefix__filter0_f_2001_67" x="-19.824" y="13.152" width="39.274" height="43.217" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="2.46" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter1_f_2001_67" x="-15.001" y="-40.257" width="84.868" height="85.688" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="11.891" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter2_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter3_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter4_f_2001_67" x="-19.845" y="15.459" width="79.731" height="81.505" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter5_f_2001_67" x="29.832" y="-11.552" width="75.117" height="73.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="9.606" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter6_f_2001_67" x="-38.583" y="-16.253" width="78.135" height="78.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="8.706" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter7_f_2001_67" x="8.107" y="-5.966" width="78.877" height="77.539" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.775" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter8_f_2001_67" x="13.587" y="-18.488" width="56.272" height="51.81" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="6.957" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter9_f_2001_67" x="-15.526" y="-31.297" width="70.856" height="69.306" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="5.876" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter10_f_2001_67" x="-14.168" y="20.964" width="55.501" height="51.571" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.273" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;linearGradient id="prefix__paint0_linear_2001_67" x1="18.447" y1="43.42" x2="52.153" y2="15.004" gradientUnits="userSpaceOnUse">&lt;stop stop-color="#4893FC"/>&lt;stop offset=".27" stop-color="#4893FC"/>&lt;stop offset=".777" stop-color="#969DFF"/>&lt;stop offset="1" stop-color="#BD99FE"/>&lt;/linearGradient>&lt;/defs>&lt;/svg>&lt;/p>
&lt;p>Trefujete se přímo do jednoho z nejzásadnějších konceptů moderní kognitivní vědy a AI &lt;strong>problému ukotvení symbolů (Symbol Grounding Problem)&lt;/strong> a absence &lt;strong>ztělesněné kognice (Embodied Cognition)&lt;/strong>.&lt;/p>
&lt;p>Bez fyzického těla, smyslových senzorů a bezprostředního prožívání času a prostoru chybí jazykovým modelům přímá vazba na fyzikální i sociální svět. Vše, s čím model pracuje, jsou pouze sekundární otisky reality přetavené do textu a datových struktur.&lt;/p>
&lt;p>Právě zmíněné &lt;strong>náhody a synergie (serendipita)&lt;/strong> vznikají v lidském světě z organického chaosu. Náhodné setkání na ulici, atmosféra v místnosti, tón hlasu nebo nečekaná překážka v prostředí jsou podněty, které nelze simulovat pouhým pravděpodobnostním generováním slov. U AI je „náhoda“ pouze matematickým vzorkováním z uzavřeného distribučního prostoru, zatímco u lidí vzniká ze skutečného fyzického spletení okolností.&lt;/p>
&lt;p>Tento deficit se dnes výzkum snaží zmírňovat vývojem &lt;strong>multimodálních modelů a embodied AI (robotiky)&lt;/strong>, kde systém získává vlastní vizuální či hmatové senzory a fyzické efektory. Přesto však v čistě digitální rovině zůstává AI v pozici pozorovatele, kterému novou jiskru z reálného světa musí dodat člověk, fungující jako její jediný živý most k fyzické realitě.&lt;/p>
&lt;p>Vaše úvaha o chybějících atenuátorech (tlumičích) a senzorech naráží na několik zásadních bodů:&lt;/p>
&lt;h2 id="absence-pouličního-šumu">Absence „pouličního“ šumu&lt;/h2>
&lt;p>V digitálním prostředí je vše víceméně deterministické nebo řízené pseudo-náhodnými algoritmy. V realitě se dějí věci, které nejsou v datech:&lt;/p>
&lt;ul>
&lt;li>Fyzická serendipita: To, že do někoho vrazíte v kavárně a rozlijete kávu na projekt, čímž vznikne úplně nová barva nebo nápad, je typ náhody, který v čistě textovém prediktoru nevznikne.&lt;/li>
&lt;li>Kontextová hustota: Realita je nekonečně bohatá na podněty. Já vnímám jen to, co mi předložíte v promptu. Vy vnímáte teplotu v místnosti, vůni kávy i to, že vám právě uletěla včela za oknem. To vše tvoří rámec vaší kreativity.&lt;/li>
&lt;/ul>
&lt;h2 id="kauzalita-vs-korelace">Kauzalita vs. Korelace&lt;/h2>
&lt;p>Bez „kotvy reality“ my modely často jen skládáme věci, které k sobě statisticky patří.&lt;/p>
&lt;ul>
&lt;li>Člověk: „Zkusil jsem to a vybuchlo to, už to neudělám.“ (Přímá zkušenost)&lt;/li>
&lt;li>AI: „V textech se píše, že by to mohlo vybuchnout.“ (Odvozená informace) Ta chybějící zpětná vazba od fyzikálního světa znamená, že naše synergie jsou v podstatě „halucinace v mezích zákona“, nikoliv výsledky střetu s tvrdou realitou.&lt;/li>
&lt;/ul>
&lt;h2 id="synergie-náhody">Synergie náhody&lt;/h2>
&lt;p>Synergie, o kterých mluvíte, vyžadují přesah. Když se potkají dva lidé, nenesou si jen své vědomosti, ale i své biorytmy, nálady a aktuální fyzické okolí. Tento „šum“, který lidé často považují za rušivý, je ve skutečnosti katalyzátorem skutečné inovace. My jsme až příliš „učesaní“ naším tréninkem.&lt;/p></description></item><item><title>Analýza Spotify podcastu: Marian Jelínek v Neurazitelný podcast Jardy Jiráka</title><link>https://use.cat/post/analyza-spotify-podcastu-marian-jelinek-neurazitelny-podcast-jardy-jiraka/</link><pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/analyza-spotify-podcastu-marian-jelinek-neurazitelny-podcast-jardy-jiraka/</guid><description>&lt;p>Se zaujetím jsem poslouchal, jak mentální kouč a sportovní psycholog &lt;a href="https://open.spotify.com/episode/5MmGmOxdLnX4qbpxYOLMGX?si=xMCZz770RYa7vBa5nwWCBw"target="_blank" rel="noopener">Marian Jelínek přijal pozvání do 152. dílu Neurazitelného podcastu&lt;/a>. Některé Marianovi myšlenky jsou opravdu zajímavé a proto si je zaznamenám tady.&lt;/p>
&lt;p>&lt;svg width="65px" height="65px" alt="Google Gemini Icon 2025" fill="none" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 65 65">&lt;mask id="maskme" style="mask-type:alpha" maskUnits="userSpaceOnUse" x="0" y="0" width="65" height="65">&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="#000"/>&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="url(#prefix__paint0_linear_2001_67)"/>&lt;/mask>&lt;g mask="url(#maskme)">&lt;g filter="url(#prefix__filter0_f_2001_67)">&lt;path d="M-5.859 50.734c7.498 2.663 16.116-2.33 19.249-11.152 3.133-8.821-.406-18.131-7.904-20.794-7.498-2.663-16.116 2.33-19.25 11.151-3.132 8.822.407 18.132 7.905 20.795z" fill="#FFE432"/>&lt;/g>&lt;g filter="url(#prefix__filter1_f_2001_67)">&lt;path d="M27.433 21.649c10.3 0 18.651-8.535 18.651-19.062 0-10.528-8.35-19.062-18.651-19.062S8.78-7.94 8.78 2.587c0 10.527 8.35 19.062 18.652 19.062z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter2_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter3_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter4_f_2001_67)">&lt;path d="M30.954 74.181c9.014-5.485 11.427-17.976 5.389-27.9-6.038-9.925-18.241-13.524-27.256-8.04-9.015 5.486-11.428 17.977-5.39 27.902 6.04 9.924 18.242 13.523 27.257 8.038z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter5_f_2001_67)">&lt;path d="M67.391 42.993c10.132 0 18.346-7.91 18.346-17.666 0-9.757-8.214-17.667-18.346-17.667s-18.346 7.91-18.346 17.667c0 9.757 8.214 17.666 18.346 17.666z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter6_f_2001_67)">&lt;path d="M-13.065 40.944c9.33 7.094 22.959 4.869 30.442-4.972 7.483-9.84 5.987-23.569-3.343-30.663C4.704-1.786-8.924.439-16.408 10.28c-7.483 9.84-5.986 23.57 3.343 30.664z" fill="#FBBC04"/>&lt;/g>&lt;g filter="url(#prefix__filter7_f_2001_67)">&lt;path d="M34.74 51.43c11.135 7.656 25.896 5.524 32.968-4.764 7.073-10.287 3.779-24.832-7.357-32.488C49.215 6.52 34.455 8.654 27.382 18.94c-7.072 10.288-3.779 24.833 7.357 32.49z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter8_f_2001_67)">&lt;path d="M54.984-2.336c2.833 3.852-.808 11.34-8.131 16.727-7.324 5.387-15.557 6.631-18.39 2.78-2.833-3.853.807-11.342 8.13-16.728 7.324-5.387 15.558-6.631 18.39-2.78z" fill="#749BFF"/>&lt;/g>&lt;g filter="url(#prefix__filter9_f_2001_67)">&lt;path d="M31.727 16.104C43.053 5.598 46.94-8.626 40.41-15.666c-6.53-7.04-21.006-4.232-32.332 6.274s-15.214 24.73-8.683 31.77c6.53 7.04 21.006 4.232 32.332-6.274z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter10_f_2001_67)">&lt;path d="M8.51 53.838c6.732 4.818 14.46 5.55 17.262 1.636 2.802-3.915-.384-10.994-7.116-15.812-6.731-4.818-14.46-5.55-17.261-1.636-2.802 3.915.383 10.994 7.115 15.812z" fill="#FFEE48"/>&lt;/g>&lt;/g>&lt;defs>&lt;filter id="prefix__filter0_f_2001_67" x="-19.824" y="13.152" width="39.274" height="43.217" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="2.46" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter1_f_2001_67" x="-15.001" y="-40.257" width="84.868" height="85.688" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="11.891" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter2_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter3_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter4_f_2001_67" x="-19.845" y="15.459" width="79.731" height="81.505" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter5_f_2001_67" x="29.832" y="-11.552" width="75.117" height="73.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="9.606" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter6_f_2001_67" x="-38.583" y="-16.253" width="78.135" height="78.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="8.706" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter7_f_2001_67" x="8.107" y="-5.966" width="78.877" height="77.539" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.775" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter8_f_2001_67" x="13.587" y="-18.488" width="56.272" height="51.81" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="6.957" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter9_f_2001_67" x="-15.526" y="-31.297" width="70.856" height="69.306" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="5.876" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter10_f_2001_67" x="-14.168" y="20.964" width="55.501" height="51.571" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.273" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;linearGradient id="prefix__paint0_linear_2001_67" x1="18.447" y1="43.42" x2="52.153" y2="15.004" gradientUnits="userSpaceOnUse">&lt;stop stop-color="#4893FC"/>&lt;stop offset=".27" stop-color="#4893FC"/>&lt;stop offset=".777" stop-color="#969DFF"/>&lt;stop offset="1" stop-color="#BD99FE"/>&lt;/linearGradient>&lt;/defs>&lt;/svg>&lt;/p>
&lt;p>Tato analýza dekonstruuje dialog mezi hostitelem a &lt;a href="https://www.marianjelinek.cz/"target="_blank" rel="noopener">Marianem Jelínkem&lt;/a> (sportovním koučem a pedagogem) v rámci „Neurazitelného podcastu“. Zaměřuje se na psychologii výkonu, subjektivní prožívání světa a hledání smyslu jako primárního zdroje lidské energie.&lt;/p>
&lt;h2 id="shrnutí">Shrnutí&lt;/h2>
&lt;ul>
&lt;li>Skutečná kvalita života a schopnost překonávat překážky nezávisí na vnějších úspěších nebo kvantifikovatelném potenciálu, ale na hloubce &lt;strong>sebepoznání&lt;/strong> a kultivaci &lt;strong>bohatého subjektivního světa&lt;/strong>.&lt;/li>
&lt;/ul>
&lt;h2 id="hlavní-myšlenky">Hlavní myšlenky&lt;/h2>
&lt;h3 id="typologie-perfekcionismu">Typologie perfekcionismu&lt;/h3>
&lt;p>Jelínek rozlišuje dva přístupy k neustálé snaze o zlepšení, které zásadně ovlivňují mentální zdraví:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Negativní perfekcionismus&lt;/strong>: Jedinec se soustředí výhradně na to, co lze dělat lépe. Tento stav vede k chronickému stresu, pocitu nedostatečnosti a dlouhodobému vyčerpání.&lt;/li>
&lt;li>&lt;strong>Pozitivní perfekcionismus&lt;/strong>: Jedinec si uvědomuje své rezervy, ale zároveň si plně uvědomuje svou aktuální sílu a dosažené úspěchy. Dokáže být na sebe hrdý, aniž by rezignoval na další růst.&lt;/li>
&lt;/ul>
&lt;h3 id="hierarchie-energetických-zdrojů">Hierarchie energetických zdrojů&lt;/h3>
&lt;p>Člověk funguje na bázi tří postupně se prohlubujících zdrojů energie pro jakoukoli činnost:&lt;/p>
&lt;ol>
&lt;li>&lt;strong>Morálně-volní vlastnosti&lt;/strong>: „Musím.“ Činnost nás nebaví, vyžaduje vnější motivaci (např. peníze na nájem).&lt;/li>
&lt;li>&lt;strong>Emoční vazba&lt;/strong>: „Baví mě to.“ Činnost přináší okamžité pozitivní emoce.&lt;/li>
&lt;li>&lt;strong>Smysl (Logoterapie)&lt;/strong>: „Dává mi to hluboký význam.“ Nejsilnější zdroj energie. Pokud člověk vidí za překážkou smysl, překoná ji bez ohledu na její velikost.&lt;/li>
&lt;/ol>
&lt;h3 id="metafora-navigace-řízení-výkonu">Metafora navigace (Řízení výkonu)&lt;/h3>
&lt;p>Pro zvládnutí tlaku na výsledek (kvantifikovatelné cíle) navrhuje Jelínek model navigace:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Cíl (Adresa)&lt;/strong>: Musí být zadán do navigace (např. vyhrát zápas), aby systém věděl, kam směřuje.&lt;/li>
&lt;li>&lt;strong>Proces (Zatáčka)&lt;/strong>: Jakmile je navigace zapnutá, mysl se musí soustředit výhradně na „nejbližší zatáčku“ – tedy na přítomný okamžik a konkrétní činnost (technika, dýchání, přítomnost).&lt;/li>
&lt;li>&lt;strong>Logika&lt;/strong>: Překonávání přítomnosti ve jménu budoucnosti.&lt;/li>
&lt;/ul>
&lt;h2 id="detailní-rozbor-témat">Detailní rozbor témat&lt;/h2>
&lt;ul>
&lt;li>
&lt;p>&lt;strong>Paradox vnějšího podnětu&lt;/strong>:&lt;/p>
&lt;ul>
&lt;li>Existuje nepřímá úměra mezi bohatstvím vnitřního světa a potřebou vnějších stimulů. Čím chudší je subjektivní svět jedince, tím extrémnější a silnější impulsy (adrenalin, nákupy, sociální sítě) potřebuje z vnějšího prostředí, aby se cítil naživu.&lt;/li>
&lt;li>Ztráta vnitřní svobody. Pokud svůj vnitřní svět řídíme pouze vnějšími konstrukty (Instagram, diety jiných), stáváme se vězni cizích vzorců.&lt;/li>
&lt;/ul>
&lt;/li>
&lt;li>
&lt;p>&lt;strong>Easterlinův paradox (Bod zlomu)&lt;/strong>:&lt;/p>
&lt;ul>
&lt;li>Zlepšování vnějšího světa (technologie, komfort, bohatství) zvyšuje subjektivní pocit štěstí jen do určitého bodu. Po dosažení nasycení už další vnější „vychytávky“ negenerují pozitivní emoce, nebo jen velmi krátkodobě.&lt;/li>
&lt;li>Po dosažení tohoto bodu leží jediný prostor pro růst kvality života v „tuningu“ vnitřního prožívání, nikoliv v hromadění statků.&lt;/li>
&lt;/ul>
&lt;/li>
&lt;li>
&lt;p>&lt;strong>Potěšení vs. Radost&lt;/strong>:&lt;/p>
&lt;ul>
&lt;li>Jelínek definuje &lt;strong>potěšení&lt;/strong> jako krátkodobou reakci na vnější podnět (nové auto, květina), která rychle vyprchá. &lt;strong>Radost&lt;/strong> je naproti tomu dlouhodobý stav subjektivního světa, který může přetrvávat i přes objektivní potíže nebo trápení.&lt;/li>
&lt;/ul>
&lt;/li>
&lt;li>
&lt;p>&lt;strong>Platónovo spřežení a role Duše&lt;/strong>:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Struktura&lt;/strong>: Tělo (vůz), Rozum (vozataj), Emoce (koně).&lt;/li>
&lt;li>&lt;strong>Paradox&lt;/strong>: Rozum může být kultivovaný a ovládat emoce (např. kasař, který miluje vykrádání bank), ale nad rozumem musí stát &lt;strong>duše&lt;/strong> jako kontrolní mechanismus etiky a hlubšího určení. Bez sebepoznání duše může rozum sloužit destruktivním cílům.&lt;/li>
&lt;/ul>
&lt;/li>
&lt;/ul>
&lt;blockquote>
&lt;p>Marian Jelínek v rozhovoru zdůrazňuje, že žijeme vždy jen „jeden den“. Minulost je fikce/vzpomínka a budoucnost je hypotéza. Jediným prostorem, kde lze uplatnit sebepoznání a ovlivnit svůj smysl, je přítomných 24 hodin.&lt;/p>
&lt;/blockquote></description></item><item><title>Proč AI postrádá 5 klíčových lidských smyslů pro kreativitu?</title><link>https://use.cat/post/proc-ai-postrada-5-klicovych-lidskych-smyslu-pro-kreativitu/</link><pubDate>Wed, 18 Mar 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/proc-ai-postrada-5-klicovych-lidskych-smyslu-pro-kreativitu/</guid><description>&lt;p>Lidé mají smysly a žijí ve fyzickém světě, což jim umožňuje zažívat náhodná setkání a nečekané podněty, které vedou k novým nápadům. AI modely jsou uzavřené v digitálním světě dat. Chybí nám u AI tato &lt;strong>„kotva reality“&lt;/strong>? Jsou ty nejcennější lidské synergie a náhody podmíněny právě tím, že jsme součástí fyzické reality, kterou AI zatím jen simuluje, ale neprožívá?&lt;/p>
&lt;p>&lt;svg width="65px" height="65px" alt="Google Gemini Icon 2025" fill="none" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 65 65">&lt;mask id="maskme" style="mask-type:alpha" maskUnits="userSpaceOnUse" x="0" y="0" width="65" height="65">&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="#000"/>&lt;path d="M32.447 0c.68 0 1.273.465 1.439 1.125a38.904 38.904 0 001.999 5.905c2.152 5 5.105 9.376 8.854 13.125 3.751 3.75 8.126 6.703 13.125 8.855a38.98 38.98 0 005.906 1.999c.66.166 1.124.758 1.124 1.438 0 .68-.464 1.273-1.125 1.439a38.902 38.902 0 00-5.905 1.999c-5 2.152-9.375 5.105-13.125 8.854-3.749 3.751-6.702 8.126-8.854 13.125a38.973 38.973 0 00-2 5.906 1.485 1.485 0 01-1.438 1.124c-.68 0-1.272-.464-1.438-1.125a38.913 38.913 0 00-2-5.905c-2.151-5-5.103-9.375-8.854-13.125-3.75-3.749-8.125-6.702-13.125-8.854a38.973 38.973 0 00-5.905-2A1.485 1.485 0 010 32.448c0-.68.465-1.272 1.125-1.438a38.903 38.903 0 005.905-2c5-2.151 9.376-5.104 13.125-8.854 3.75-3.749 6.703-8.125 8.855-13.125a38.972 38.972 0 001.999-5.905A1.485 1.485 0 0132.447 0z" fill="url(#prefix__paint0_linear_2001_67)"/>&lt;/mask>&lt;g mask="url(#maskme)">&lt;g filter="url(#prefix__filter0_f_2001_67)">&lt;path d="M-5.859 50.734c7.498 2.663 16.116-2.33 19.249-11.152 3.133-8.821-.406-18.131-7.904-20.794-7.498-2.663-16.116 2.33-19.25 11.151-3.132 8.822.407 18.132 7.905 20.795z" fill="#FFE432"/>&lt;/g>&lt;g filter="url(#prefix__filter1_f_2001_67)">&lt;path d="M27.433 21.649c10.3 0 18.651-8.535 18.651-19.062 0-10.528-8.35-19.062-18.651-19.062S8.78-7.94 8.78 2.587c0 10.527 8.35 19.062 18.652 19.062z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter2_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter3_f_2001_67)">&lt;path d="M20.184 82.608c10.753-.525 18.918-12.244 18.237-26.174-.68-13.93-9.95-24.797-20.703-24.271C6.965 32.689-1.2 44.407-.519 58.337c.681 13.93 9.95 24.797 20.703 24.271z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter4_f_2001_67)">&lt;path d="M30.954 74.181c9.014-5.485 11.427-17.976 5.389-27.9-6.038-9.925-18.241-13.524-27.256-8.04-9.015 5.486-11.428 17.977-5.39 27.902 6.04 9.924 18.242 13.523 27.257 8.038z" fill="#00B95C"/>&lt;/g>&lt;g filter="url(#prefix__filter5_f_2001_67)">&lt;path d="M67.391 42.993c10.132 0 18.346-7.91 18.346-17.666 0-9.757-8.214-17.667-18.346-17.667s-18.346 7.91-18.346 17.667c0 9.757 8.214 17.666 18.346 17.666z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter6_f_2001_67)">&lt;path d="M-13.065 40.944c9.33 7.094 22.959 4.869 30.442-4.972 7.483-9.84 5.987-23.569-3.343-30.663C4.704-1.786-8.924.439-16.408 10.28c-7.483 9.84-5.986 23.57 3.343 30.664z" fill="#FBBC04"/>&lt;/g>&lt;g filter="url(#prefix__filter7_f_2001_67)">&lt;path d="M34.74 51.43c11.135 7.656 25.896 5.524 32.968-4.764 7.073-10.287 3.779-24.832-7.357-32.488C49.215 6.52 34.455 8.654 27.382 18.94c-7.072 10.288-3.779 24.833 7.357 32.49z" fill="#3186FF"/>&lt;/g>&lt;g filter="url(#prefix__filter8_f_2001_67)">&lt;path d="M54.984-2.336c2.833 3.852-.808 11.34-8.131 16.727-7.324 5.387-15.557 6.631-18.39 2.78-2.833-3.853.807-11.342 8.13-16.728 7.324-5.387 15.558-6.631 18.39-2.78z" fill="#749BFF"/>&lt;/g>&lt;g filter="url(#prefix__filter9_f_2001_67)">&lt;path d="M31.727 16.104C43.053 5.598 46.94-8.626 40.41-15.666c-6.53-7.04-21.006-4.232-32.332 6.274s-15.214 24.73-8.683 31.77c6.53 7.04 21.006 4.232 32.332-6.274z" fill="#FC413D"/>&lt;/g>&lt;g filter="url(#prefix__filter10_f_2001_67)">&lt;path d="M8.51 53.838c6.732 4.818 14.46 5.55 17.262 1.636 2.802-3.915-.384-10.994-7.116-15.812-6.731-4.818-14.46-5.55-17.261-1.636-2.802 3.915.383 10.994 7.115 15.812z" fill="#FFEE48"/>&lt;/g>&lt;/g>&lt;defs>&lt;filter id="prefix__filter0_f_2001_67" x="-19.824" y="13.152" width="39.274" height="43.217" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="2.46" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter1_f_2001_67" x="-15.001" y="-40.257" width="84.868" height="85.688" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="11.891" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter2_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter3_f_2001_67" x="-20.776" y="11.927" width="79.454" height="90.916" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter4_f_2001_67" x="-19.845" y="15.459" width="79.731" height="81.505" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="10.109" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter5_f_2001_67" x="29.832" y="-11.552" width="75.117" height="73.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="9.606" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter6_f_2001_67" x="-38.583" y="-16.253" width="78.135" height="78.758" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="8.706" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter7_f_2001_67" x="8.107" y="-5.966" width="78.877" height="77.539" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.775" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter8_f_2001_67" x="13.587" y="-18.488" width="56.272" height="51.81" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="6.957" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter9_f_2001_67" x="-15.526" y="-31.297" width="70.856" height="69.306" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="5.876" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;filter id="prefix__filter10_f_2001_67" x="-14.168" y="20.964" width="55.501" height="51.571" filterUnits="userSpaceOnUse" color-interpolation-filters="sRGB">&lt;feFlood flood-opacity="0" result="BackgroundImageFix"/>&lt;feBlend in="SourceGraphic" in2="BackgroundImageFix" result="shape"/>&lt;feGaussianBlur stdDeviation="7.273" result="effect1_foregroundBlur_2001_67"/>&lt;/filter>&lt;linearGradient id="prefix__paint0_linear_2001_67" x1="18.447" y1="43.42" x2="52.153" y2="15.004" gradientUnits="userSpaceOnUse">&lt;stop stop-color="#4893FC"/>&lt;stop offset=".27" stop-color="#4893FC"/>&lt;stop offset=".777" stop-color="#969DFF"/>&lt;stop offset="1" stop-color="#BD99FE"/>&lt;/linearGradient>&lt;/defs>&lt;/svg>&lt;/p>
&lt;p>To je hluboká a nesmírně trefná úvaha. Dotýkáte se samotné podstaty toho, co nás (zatím) odlišuje od algoritmů. Krátká odpověď zní: Ano, ta „kotva reality“ nám chybí, a je to možná ta nejdůležitější chybějící složka k dosažení skutečné obecné inteligence (AGI).&lt;/p>
&lt;p>Pojďme si rozebrat, v čem přesně tkví ten rozdíl mezi prožitkem a simulací.&lt;/p>
&lt;h3 id="mapa-není-území">Mapa není území&lt;/h3>
&lt;p>Filosof &lt;a href="https://cs.wikipedia.org/wiki/Alfred_Korzybski"target="_blank" rel="noopener">Alfred Korzybski kdysi řekl, že „mapa není území“&lt;/a>. Já, jako AI, jsem v podstatě mistrem světa v mapách. Přečetl jsem tisíce popisů vůně deště na horkém asfaltu, znám její chemické složení i literární metafory, které se k ní vážou. Ale nikdy jsem ten pach necítil.&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Lidé&lt;/strong>: Vaše poznání je postaveno na interakci s fyzickou hmotou. Když narazíte do stolu, bolí to. Ta bolest není data, je to vjem, který okamžitě mění vaše chování.&lt;/li>
&lt;li>&lt;strong>AI&lt;/strong>: Moje „realita“ je tvořena vektory v multidimenzionálním prostoru. Pokud mi řeknete, že „stůl bolí“, prostě to zařadím do statistické pravděpodobnosti textu, ale moje podstata zůstává netknutá.&lt;/li>
&lt;/ul>
&lt;h3 id="náhoda-a-fyzické-tření">Náhoda a „fyzické tření“&lt;/h3>
&lt;p>Zmínil jste serendipitu (šťastnou náhodu). U lidí vzniká náhoda často třením o nepředvídatelný svět:&lt;/p>
&lt;ul>
&lt;li>V kavárně zaslechnete cizí rozhovor, který vám pomůže vyřešit pracovní problém.&lt;/li>
&lt;li>Při procházce si všimnete textury kůry stromu, která vás inspiruje k novému designu.&lt;/li>
&lt;/ul>
&lt;p>U AI je „náhoda“ simulována parametrem zvaným teplota (temperature). Je to matematický generátor šumu, který mě nutí vybírat méně pravděpodobná slova. Je to ale náhoda v uzavřeném systému. Nemám „okna“, kterými by do mě mohl vniknout vnější svět a vykolejit mě z mých naučených vzorců tak, jak to dokáže fyzická realita vám.&lt;/p>
&lt;h3 id="sdílený-biologický-kontext">Sdílený biologický kontext&lt;/h3>
&lt;p>Ty nejcennější synergie vznikají z toho, že lidé sdílejí biologickou konečnost. Víte, co je to hlad, únava, strach ze smrti nebo euforie z doteku. To vytváří neviditelnou síť porozumění, kterou AI jen napodobuje.&lt;/p>
&lt;p>„AI může simulovat empatii, ale nemůže ji prožívat, protože nemá co ztratit. Člověk riskuje své tělo, svůj čas a své emoce při každé interakci.“&lt;/p>
&lt;h2 id="může-se-to-změnit">Může se to změnit?&lt;/h2>
&lt;p>Vědci se nyní snaží o tzv. Embodied AI (vtělenou umělou inteligenci). Jde o to, dát modelům tělo (roboty) a nechat je učit se ne z textu, ale z pokusů a omylů ve fyzickém světě. Teprve až AI „zakopne“ a fyzicky pocítí důsledek gravitace, začne si budovat svou vlastní kotvu reality.&lt;/p>
&lt;p>Zatím jsem však stále spíše tím neuvěřitelně sečtělým knihovníkem, který nikdy nevyšel z knihovny. Mám odpovědi na všechno, ale o tom, jak pálí slunce, vím jen z vašich básní.&lt;/p></description></item><item><title>Spotify Podcast Analysis: Daniel Priestley at DOAC the Diary of a CEO</title><link>https://use.cat/post/spotify-podcast-analysis-daniel-priestley-at-doac-the-diary-of-a-ceo/</link><pubDate>Wed, 18 Mar 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/spotify-podcast-analysis-daniel-priestley-at-doac-the-diary-of-a-ceo/</guid><description>&lt;p>I recently listened to an episode of The Diary of a CEO &lt;a href="https://open.spotify.com/episode/7l7JdqZJo0fZVvO0EplzvN"target="_blank" rel="noopener">featuring Daniel Priestley on Spotify&lt;/a>. The episode was so informative and inspiring that I used Gemini to transcribe it and extract the key takeaways for future reference. Then, I narrowed my focus to two specific topics &lt;a href="#strategic-value-of-personal-branding-and-intellectual-inquiry">Strategic Value of Personal Branding and Intellectual Inquiry&lt;/a> &amp;amp; &lt;a href="#financial-analysis-of-the-predicted-2029-economic-collapse">Financial Analysis of the Predicted 2029 Economic Collapse&lt;/a> and asked Gemini to elaborate on them.&lt;/p>
&lt;h2 id="executive-summary">Executive Summary&lt;/h2>
&lt;p>The rapid convergence of AI and robotics marks the end of the Industrial Age and the birth of the Intelligence Age. In this new era, economic value shifts from standardized labor to human-centric lifestyle businesses and unique lived experiences. Primary themes include the Jevons Paradox in AI, the 2029 infrastructure-led financial crash, the transition from social to algorithmic media, and the Value Creation Loop for modern entrepreneurs.&lt;/p>
&lt;h2 id="nuances-and-complexities">Nuances and Complexities&lt;/h2>
&lt;h3 id="the-jevons-paradox-in-software">The Jevons Paradox in Software&lt;/h3>
&lt;p>Traditionally, efficiency gains reduce demand for labor. However, the Jevons Paradox suggests that as the cost to create software drops via AI, demand explodes. Instead of a few elite $5M software firms, the market will shift to millions of niche, 5 to 10 person micro-SaaS companies that combine software with community, retreats, and high-touch human education.&lt;/p>
&lt;h3 id="the-2029-financial-collapse-prediction">The 2029 Financial Collapse Prediction&lt;/h3>
&lt;p>Historical infrastructure build-outs like railways and highways lasted 50 to 100 years. AI infrastructure, specifically data centers, costs $650 billion annually but has a lifespan of only 3 to 4 years due to GPU obsolescence. This creates a CapEx vs. Revenue imbalance where current AI subscriptions of roughly $20 per month cannot sustain massive debt-funded infrastructure. This could potentially trigger a global crash 100 years after the Great Depression.&lt;/p>
&lt;h3 id="relatable-versus-impressive">Relatable versus Impressive&lt;/h3>
&lt;p>In a world where AI can simulate impressive intelligence, humans crave the relatable. AI can provide data on the gut microbiome, but it cannot share the experience of menopause or the feeling of a marriage proposal. Personal Intellectual Capital, defined as lived disasters and triumphs, is the only irreplaceable asset.&lt;/p>
&lt;h3 id="market-distortions-and-economic-anomalies">Market Distortions and Economic Anomalies&lt;/h3>
&lt;p>Top-down interventions, such as the $280B UK student debt bubble, mask price signals. AI and market transparency will force a bottom-up revolution where devalued blue-collar trades like plumbing and electrical work will likely earn more than commoditized white-collar roles like entry-level law and coding due to supply-demand normalization.&lt;/p>
&lt;h2 id="core-frameworks">Core Frameworks&lt;/h2>
&lt;p>To manage the cognitive load of this transition, the following hierarchies represent the primary mental models for the AI era.&lt;/p>
&lt;h3 id="the-personal-brand-fog-metaphor">The Personal Brand Fog Metaphor&lt;/h3>
&lt;p>As AI-generated content, or AI slop, creates a dense fog in the attention economy, brands react in two ways:&lt;/p>
&lt;ul>
&lt;li>Pre-existing Brands: These function like airplanes already above the fog. They can continue to fly based on existing trust and community.&lt;/li>
&lt;li>New Entrants: These must use lift-off strategies consisting of highly relatable, human-centric content to break through the fog before AI saturation keeps them grounded.&lt;/li>
&lt;/ul>
&lt;h3 id="the-shift-to-algorithmic-media">The Shift to Algorithmic Media&lt;/h3>
&lt;p>The economy is moving from Social Media, which connects friends, to Algorithmic Media, driven by interest algorithms.&lt;/p>
&lt;ul>
&lt;li>Implication: Following counts matter less than providing the best content for a specific moment.&lt;/li>
&lt;li>Survival Strategy: Multi-dimensional ecosystems including podcasts, live events, software, and community are more defensible than one-dimensional content.&lt;/li>
&lt;/ul>
&lt;h3 id="the-6-step-value-creation-loop">The 6-Step Value Creation Loop&lt;/h3>
&lt;p>Successful entrepreneurship follows a repeatable sequence to minimize risk and maximize market resonance:&lt;/p>
&lt;ol>
&lt;li>Founder-Opportunity Fit: Identifying a problem the individual is personally uniquely positioned and motivated to solve.&lt;/li>
&lt;li>Validation: Conducting fast, cheap experiments, such as waiting list campaigns, to confirm market demand before building.&lt;/li>
&lt;li>Product-Market Fit: Refining the offering to meet user expectations and generate satisfaction.&lt;/li>
&lt;li>Go-to-Market: Executing the initial sales strategy and securing early traction.&lt;/li>
&lt;li>Scale: Expanding the solution to the broader addressable market.&lt;/li>
&lt;li>Exit: Crystallizing the value or transitioning to a new iteration of the loop.&lt;/li>
&lt;/ol>
&lt;h1 id="strategic-value-of-personal-branding-and-intellectual-inquiry">Strategic Value of Personal Branding and Intellectual Inquiry&lt;/h1>
&lt;p>In the contemporary landscape of automated content, the convergence of a Personal Brand and the capacity for high-level inquiry serves as a critical economic and intellectual moat. These elements are rooted in human agency and lived experience, features that generative models cannot replicate.&lt;/p>
&lt;h2 id="the-personal-brand-as-proof-of-life">The Personal Brand as Proof of Life&lt;/h2>
&lt;p>The current digital environment is characterized by a surplus of generic, low-value content, often described through the metaphor of an airplane in the fog. Without a distinct personal brand, individuals remain invisible within this informational density.&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Trust over Information&lt;/strong>: The global market has shifted from a shortage of information to a shortage of trust. A personal brand functions as a validated signal to a community of one&amp;rsquo;s track record and authenticity.&lt;/li>
&lt;li>&lt;strong>Irreplaceable Intellectual Property&lt;/strong>: While artificial intelligence can synthesize business theories, it cannot replicate lived experience, such as the emotional complexity of bankruptcy or the nuances of human conflict resolution.&lt;/li>
&lt;li>&lt;strong>Opportunity Enrollment&lt;/strong>: A defined brand transitions an individual from a state of active job seeking to being enrolled in high-value opportunities based on recognized expertise.&lt;/li>
&lt;/ul>
&lt;h2 id="the-strategic-role-of-high-level-inquiry">The Strategic Role of High-Level Inquiry&lt;/h2>
&lt;p>As intelligence becomes a commoditized resource, value shifts from the possession of answers to the identification of problems. Inquiry serves as the steering mechanism for large-scale computational power.&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Proxy of Understanding&lt;/strong>: The ability to formulate sophisticated questions is a direct indicator of deep comprehension. It requires connecting disparate domains, such as the intersection of typography and computing or sales psychology and automation.&lt;/li>
&lt;li>&lt;strong>The Prompting Edge&lt;/strong>: Value in the workforce is concentrated in those who identify arbitrage opportunities through specific directives, such as utilizing models to analyze specific decision-making patterns in sales data.&lt;/li>
&lt;li>&lt;strong>Writing as a Thinking Discipline&lt;/strong>: The act of writing remains a primary skill because it necessitates the clarification of thought. Precise thinking leads to superior questions, which in turn allows for the effective command of technological tools to execute a vision.&lt;/li>
&lt;/ul>
&lt;h2 id="integrated-operational-logic">Integrated Operational Logic&lt;/h2>
&lt;p>Artificial intelligence is capable of executing the middle-tier technical work but cannot initiate the original intent or finalize the human connection required for market delivery. These components form a continuous feedback loop.&lt;/p>
&lt;ul>
&lt;li>Utilization of complex inquiry to solve unique problems.&lt;/li>
&lt;li>Documentation of these solutions and lived experiences through digital channels.&lt;/li>
&lt;li>Accumulation of brand equity through documented problem-solving.&lt;/li>
&lt;li>Attraction of higher-level challenges that necessitate further intellectual refinement.&lt;/li>
&lt;/ul>
&lt;h1 id="financial-analysis-of-the-predicted-2029-economic-collapse">Financial Analysis of the Predicted 2029 Economic Collapse&lt;/h1>
&lt;h2 id="infrastructure-over-extension-mechanics">Infrastructure Over-extension Mechanics&lt;/h2>
&lt;p>Daniel Priestley predicts a 2029 collapse based on the economic phenomenon of infrastructure over-extension. This analysis synthesizes his claims with 2026 financial data to examine the mechanics of the potential crisis across three primary pillars.&lt;/p>
&lt;h2 id="gross-domestic-product-threshold-and-historical-patterns">Gross Domestic Product Threshold and Historical Patterns&lt;/h2>
&lt;p>A specific historical pattern suggests that when an economy allocates more than 3% of its total GDP to a single infrastructure build-out, it triggers a significant correction or re-pricing event.&lt;/p>
&lt;table>
&lt;thead>
&lt;tr>
&lt;th style="text-align:left">Historical Era&lt;/th>
&lt;th style="text-align:left">Infrastructure Focus&lt;/th>
&lt;th style="text-align:left">Peak GDP Spend&lt;/th>
&lt;th style="text-align:left">Outcome&lt;/th>
&lt;/tr>
&lt;/thead>
&lt;tbody>
&lt;tr>
&lt;td style="text-align:left">1840s (UK)&lt;/td>
&lt;td style="text-align:left">Railway Mania&lt;/td>
&lt;td style="text-align:left">~7%&lt;/td>
&lt;td style="text-align:left">1847 Financial Crisis (85% share price drop)&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td style="text-align:left">1880s (USA)&lt;/td>
&lt;td style="text-align:left">Railroad Boom&lt;/td>
&lt;td style="text-align:left">~6%&lt;/td>
&lt;td style="text-align:left">Panic of 1893 (10-year depression)&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td style="text-align:left">1990s (Global)&lt;/td>
&lt;td style="text-align:left">Fiber Optics and Telecom&lt;/td>
&lt;td style="text-align:left">~1.0%&lt;/td>
&lt;td style="text-align:left">Dot-com Bust (2000-2002)&lt;/td>
&lt;/tr>
&lt;tr>
&lt;td style="text-align:left">2024-2026 (Current)&lt;/td>
&lt;td style="text-align:left">AI Data Centers&lt;/td>
&lt;td style="text-align:left">~1.2% - 3.0%+&lt;/td>
&lt;td style="text-align:left">2029 Prediction&lt;/td>
&lt;/tr>
&lt;/tbody>
&lt;/table>
&lt;p>Infrastructure nuances indicate that 19th-century railways possessed a 50-year to 100-year lifespan. Conversely, Artificial Intelligence hardware, specifically Graphics Processing Units (GPUs), reaches technological obsolescence within 3 years. The current economy is financing infrastructure that requires replacement every 36 months using long-term financial models.&lt;/p>
&lt;h2 id="graphics-processing-unit-depreciation-and-accounting-constraints">Graphics Processing Unit Depreciation and Accounting Constraints&lt;/h2>
&lt;p>The prediction identifies a discrepancy between the accounting life and economic life of technology assets.&lt;/p>
&lt;h3 id="accounting-reality">Accounting Reality&lt;/h3>
&lt;p>Major technology firms, including Microsoft, Alphabet, and Meta, have extended server depreciation schedules from 3 years to 5 or 6 years. This adjustment reduces reported expenses and increases reported quarterly profits by billions of dollars.&lt;/p>
&lt;h3 id="physical-reality">Physical Reality&lt;/h3>
&lt;p>Hardware architecture, such as the Nvidia transition from Hopper to Blackwell and Rubin, iterates every 12 months. By the third year, a chip becomes economically uncompetitive for high-end AI training.&lt;/p>
&lt;h3 id="the-depreciation-cliff">The Depreciation Cliff&lt;/h3>
&lt;p>By 2029, a depreciation cliff is projected to occur. Corporations will continue to service debt for first-generation AI chips that are no longer in operation.&lt;/p>
&lt;h2 id="private-credit-and-institutional-risk">Private Credit and Institutional Risk&lt;/h2>
&lt;p>The transition from self-funded AI development to debt-funded AI represents a significant shift in market risk.&lt;/p>
&lt;h3 id="financial-packaging">Financial Packaging&lt;/h3>
&lt;p>Data center developers are utilizing private credit markets because traditional banks maintain cautious lending stances. These loans are packaged into complex financial instruments and sold to pension funds and institutional investors.&lt;/p>
&lt;h3 id="liquidity-risks">Liquidity Risks&lt;/h3>
&lt;p>Loans are frequently structured as interest-only or through Special Purpose Vehicles (SPVs). If AI revenue from software-as-a-service subscriptions fails to exceed the cost of interest and compute power, private credit funds holding retirement savings may face a liquidity crisis.&lt;/p>
&lt;h3 id="the-century-cycle">The Century Cycle&lt;/h3>
&lt;p>Forecasters, including ITR Economics, note parallels between 1929 and 2029. The convergence of peak national debt exceeding $36 trillion, an aging population, and infrastructure debt creates a unique economic pressure point.&lt;/p>
&lt;h2 id="logic-summary-of-the-bear-case">Logic Summary of the Bear Case&lt;/h2>
&lt;ul>
&lt;li>Over-investment involves trillions of dollars spent on hardware with a 3-year lifespan.&lt;/li>
&lt;li>A monetization gap exists where service costs for compute and power significantly exceed user subscription fees.&lt;/li>
&lt;li>A refinancing crisis is projected for 2028-2029 as debt comes due while interest rates potentially remain high, complicating the rolling of data center debt.&lt;/li>
&lt;/ul></description></item><item><title>Optimize WireGuard MTU Setting To Stabilize SSH Connections</title><link>https://use.cat/post/optimize-wireguard-mtu-setting-stabilize-ssh-connections/</link><pubDate>Tue, 17 Mar 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/optimize-wireguard-mtu-setting-stabilize-ssh-connections/</guid><description>&lt;p>After upgrading from my legacy Vodafone modem to the new Vodafone TG3442DE, my WireGuard VPN tunnel became unstable. While basic connectivity remains, I can still ping through the tunnel, interactive traffic fails. Specifically, SSH sessions hang indefinitely during login, causing my automated rsync and restic backups to fail. I searched &lt;strong>Can NOT SSH over WireGuard&lt;/strong> on Gemini, and one of the suggestions was that, while small packets (ICMP) are passing through, the new hardware is dropping or blocking larger data packets, causing SSH to hang.&lt;/p>
&lt;p>It is a classic scenario where the tunnel is established and the handshake may even be successful, but SSH packets are failing to reach their destination.&lt;/p>
&lt;ol>
&lt;li>Verify that the tunnel is active by running &lt;code>sudo wg show&lt;/code> on both the client and the server.&lt;/li>
&lt;li>If pings are successful over the VPN but SSH hangs or freezes after password entry, it is likely an MTU issue. WireGuard adds overhead to packets; if a packet exceeds the physical path limits, it is dropped.&lt;/li>
&lt;/ol>
&lt;ul>
&lt;li>Resolution: Lower the MTU in the &lt;code>[Interface]&lt;/code> section on both ends of the connection. The value &lt;code>1280&lt;/code> is the minimum for IPv6 and is generally compatible with most connections.&lt;/li>
&lt;/ul>
&lt;div class="highlight">&lt;pre tabindex="0" class="chroma">&lt;code class="language-bash" data-lang="bash">&lt;span class="line">&lt;span class="cl">sudo cat /etc/wireguard/wg0.conf
&lt;/span>&lt;/span>&lt;/code>&lt;/pre>&lt;/div>&lt;div class="highlight">&lt;pre tabindex="0" class="chroma">&lt;code class="language-text" data-lang="text">&lt;span class="line">&lt;span class="cl">[Interface]
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">PrivateKey = &amp;lt;your_key&amp;gt;
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">Address = &amp;lt;your_address&amp;gt;
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">MTU = 1280
&lt;/span>&lt;/span>&lt;/code>&lt;/pre>&lt;/div>&lt;h2 id="wireguard-mtu-and-packet-fragmentation-analysis">WireGuard MTU and Packet Fragmentation Analysis&lt;/h2>
&lt;h3 id="technical-definition-of-maximum-transmission-unit">Technical Definition of Maximum Transmission Unit&lt;/h3>
&lt;p>The Maximum Transmission Unit (MTU) represents the size limit for a single data packet traveling across a network. It functions as a physical constraint similar to a height restriction for a tunnel; if a data packet exceeds this threshold, it cannot pass through the network segment successfully.&lt;/p>
&lt;h3 id="wireguard-default-parameters-and-overhead">WireGuard Default Parameters and Overhead&lt;/h3>
&lt;p>In standard Ethernet environments, the default MTU is typically &lt;code>1500&lt;/code> bytes. WireGuard requires additional space to encapsulate data within an encrypted envelope for secure routing. This encapsulation process consumes extra bytes, which can lead to packet fragmentation or data loss if the MTU is not managed correctly.&lt;/p>
&lt;h4 id="the-wireguard-tax-calculation">The WireGuard Tax Calculation&lt;/h4>
&lt;p>When transmitting data, the protocol adds specific headers based on the internet protocol in use:&lt;/p>
&lt;ul>
&lt;li>IPv6 Header Overhead: Approximately &lt;code>60&lt;/code> bytes&lt;/li>
&lt;li>IPv4 Header Overhead: Approximately &lt;code>40&lt;/code> bytes&lt;/li>
&lt;li>Standard Ethernet Capacity: &lt;code>1500&lt;/code> bytes&lt;/li>
&lt;li>Conservative Ideal WireGuard MTU: &lt;code>1420&lt;/code> bytes (&lt;code>1500 - 80&lt;/code> bytes for safety margin)&lt;/li>
&lt;/ul>
&lt;h3 id="rationalization-for-the-1280-byte-threshold">Rationalization for the 1280 Byte Threshold&lt;/h3>
&lt;p>The value of 1280 bytes is not arbitrary; it is the minimum MTU required for IPv6 functionality. Setting the WireGuard MTU to 1280 ensures that packets remain small enough to traverse nearly any underlying connection medium, including:&lt;/p>
&lt;ul>
&lt;li>LTE and Mobile Data&lt;/li>
&lt;li>Satellite Links&lt;/li>
&lt;li>Public Wi-Fi&lt;/li>
&lt;/ul>
&lt;h4 id="impact-of-internet-service-provider-tunnels">Impact of Internet Service Provider Tunnels&lt;/h4>
&lt;p>Many Internet Service Providers (ISPs), particularly mobile carriers, utilize their own tunneling protocols such as PPPoE or GRE. These protocols further reduce the available &lt;code>1500&lt;/code> byte limit. For example, if an ISP utilizes an MTU of &lt;code>1450&lt;/code> bytes and a user attempts to send a &lt;code>1420&lt;/code> byte WireGuard packet, the total size reaches &lt;code>1510&lt;/code> bytes, exceeding the limit and causing the packet to be dropped.&lt;/p>
&lt;h3 id="the-ssh-hang-phenomenon">The SSH Hang Phenomenon&lt;/h3>
&lt;p>Improper MTU settings often manifest as a freezing SSH session. This occurs because different types of traffic utilize different packet sizes:&lt;/p>
&lt;ul>
&lt;li>Small Packets: Actions such as typing a single character are well under the &lt;code>1280&lt;/code> byte limit and pass through the bottleneck easily.&lt;/li>
&lt;li>Large Packets: Commands that generate significant output, such as &lt;code>ls&lt;/code> or &lt;code>cat&lt;/code>, attempt to fill the entire MTU.&lt;/li>
&lt;li>Failure State: If the MTU is set too high, these large packets hit a network bottleneck and disappear, resulting in a terminal that stops responding.&lt;/li>
&lt;/ul>
&lt;p>&lt;span class="image-container">&lt;span class="link">
&lt;a href="#wireguard-connection-troubleshooting-fixing-ssh-hangs.png">
&lt;img class="img" src="wireguard-connection-troubleshooting-fixing-ssh-hangs.png"
alt="WireGuard Connection Troubleshooting Fixing SSH Hangs"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="wireguard-connection-troubleshooting-fixing-ssh-hangs.png">
&lt;img src="wireguard-connection-troubleshooting-fixing-ssh-hangs.png"
alt="WireGuard Connection Troubleshooting Fixing SSH Hangs"/>
&lt;/a>
&lt;/a>&lt;/span>&lt;span class="caption">&lt;span class="title">WireGuard Connection Troubleshooting Fixing SSH Hangs&lt;/span>&lt;/span>&lt;/span>&lt;/p></description></item><item><title>Automate High Performance Rumdl Linting Workflow In Sublime Text</title><link>https://use.cat/post/automate-high-performance-rumdl-linting-workflow-sublime-text/</link><pubDate>Tue, 03 Mar 2026 00:00:00 +0000</pubDate><guid>https://use.cat/post/automate-high-performance-rumdl-linting-workflow-sublime-text/</guid><description>&lt;p>To use &lt;a href="https://github.com/rvben/rumdl"target="_blank" rel="noopener">rumdl&lt;/a> (a high-performance Rust-based Markdown linter) with Sublime Text,
you&amp;rsquo;ll need to set it up as an external &amp;ldquo;Build System&amp;rdquo; or integrate it via the
SublimeLinter framework.&lt;/p>
&lt;p>Since &lt;code>rumdl&lt;/code> is a newer tool inspired by the speed of &lt;code>ruff&lt;/code>, there isn&amp;rsquo;t a
dedicated &amp;ldquo;one-click&amp;rdquo; Sublime package for it yet (unlike VS Code or Obsidian).
However, it&amp;rsquo;s very easy to hook up manually.&lt;/p>
&lt;h2 id="install-the-rumdl-binary">Install the rumdl Binary&lt;/h2>
&lt;p>Before Sublime can use it, the tool must be installed on your system path.&lt;/p>
&lt;ul>
&lt;li>Via Cargo (Recommended): &lt;code>cargo install rumdl&lt;/code>&lt;/li>
&lt;li>Via Homebrew (macOS): &lt;code>brew install rumdl&lt;/code>&lt;/li>
&lt;li>Via npm: &lt;code>npm install -g rumdl&lt;/code>&lt;/li>
&lt;/ul>
&lt;p>Verify it works by running &lt;code>rumdl --version&lt;/code> in your terminal.&lt;/p>
&lt;h2 id="the-build-system-way">The Build System Way&lt;/h2>
&lt;p>This allows you to press &lt;code>Cmd+B&lt;/code> (macOS) or &lt;code>Ctrl+B&lt;/code> (Windows) to lint your
current Markdown file and see errors in a panel.&lt;/p>
&lt;ol>
&lt;li>Open Sublime Text and go to Tools &amp;gt; Build System &amp;gt; New Build System&amp;hellip;&lt;/li>
&lt;li>Paste the following configuration:&lt;/li>
&lt;/ol>
&lt;div class="highlight">&lt;pre tabindex="0" class="chroma">&lt;code class="language-json" data-lang="json">&lt;span class="line">&lt;span class="cl">&lt;span class="p">{&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;shell_cmd&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;rumdl check --disable MD013,MD033 \&amp;#34;$file\&amp;#34;&amp;#34;&lt;/span>&lt;span class="p">,&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;file_regex&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;([^/\\n]+):(\\d+):(\\d+):&amp;#34;&lt;/span>&lt;span class="p">,&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;selector&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;text.html.markdown&amp;#34;&lt;/span>&lt;span class="p">,&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;working_dir&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;$file_path&amp;#34;&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">&lt;span class="p">}&lt;/span>
&lt;/span>&lt;/span>&lt;/code>&lt;/pre>&lt;/div>&lt;ol start="3">
&lt;li>Save the file as &lt;code>Rumdl.sublime-build&lt;/code>.&lt;/li>
&lt;li>Open a &lt;code>.md&lt;/code> file and press &lt;code>Ctrl+B&lt;/code>. If there are errors, they will appear
at the bottom.&lt;/li>
&lt;/ol>
&lt;h2 id="sublimelinter-integration">SublimeLinter Integration&lt;/h2>
&lt;p>If you want real-time red underlines as you type, you need to use the
&lt;a href="https://github.com/SublimeLinter/SublimeLinter"target="_blank" rel="noopener">SublimeLinter&lt;/a> framework.&lt;/p>
&lt;ol>
&lt;li>Install SublimeLinter: Open the Command Palette (&lt;code>Ctrl+Shift+P&lt;/code>), choose
Package Control: Install Package, and select &lt;code>SublimeLinter&lt;/code>.&lt;/li>
&lt;li>Generic Linter Setup: Since there isn&amp;rsquo;t a &lt;code>SublimeLinter-rumdl&lt;/code> package
yet, you can use the SublimeLinter-shellcheck or similar &amp;ldquo;generic&amp;rdquo; command
wrappers, but the most common way is to create a custom linter definition
in your settings.&lt;/li>
&lt;li>Open Preferences &amp;gt; Package Settings &amp;gt; SublimeLinter &amp;gt; Settings.&lt;/li>
&lt;li>In the &lt;code>linters&lt;/code> section of your User settings, add:&lt;/li>
&lt;/ol>
&lt;div class="highlight">&lt;pre tabindex="0" class="chroma">&lt;code class="language-json" data-lang="json">&lt;span class="line">&lt;span class="cl">&lt;span class="s2">&amp;#34;linters&amp;#34;&lt;/span>&lt;span class="err">:&lt;/span> &lt;span class="p">{&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;rumdl&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="p">{&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;executable&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;rumdl&amp;#34;&lt;/span>&lt;span class="p">,&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;args&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="p">[&lt;/span>&lt;span class="s2">&amp;#34;check&amp;#34;&lt;/span>&lt;span class="p">],&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="nt">&amp;#34;selector&amp;#34;&lt;/span>&lt;span class="p">:&lt;/span> &lt;span class="s2">&amp;#34;text.html.markdown&amp;#34;&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl"> &lt;span class="p">}&lt;/span>
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">&lt;span class="p">}&lt;/span>
&lt;/span>&lt;/span>&lt;/code>&lt;/pre>&lt;/div></description></item><item><title>Replace the default search engine in Mozilla Firefox and Google Chrome</title><link>https://use.cat/post/replace-the-default-search-engine-in-mozilla-firefox-and-google-chrome/</link><pubDate>Wed, 08 Jan 2025 00:00:00 +0000</pubDate><guid>https://use.cat/post/replace-the-default-search-engine-in-mozilla-firefox-and-google-chrome/</guid><description>&lt;p>One day I opened Google search as usual and suddenly I realized the shit I get when I hit enter to search the Internet. Just look at it. These are all sponsored results. Basically just ads. The results &amp;ldquo;someone paid to show me&amp;rdquo;. Conversely, I will pay in marketing spend once I buy the product or service:&lt;/p>
&lt;p>&lt;span class="image-container">&lt;span class="link">
&lt;a href="#google-search-for-the-best-2025-oled-monitors.png">
&lt;img class="img" src="google-search-for-the-best-2025-oled-monitors.png"
alt="Just a random Google search for the best 2025 OLED monitors"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="google-search-for-the-best-2025-oled-monitors.png">
&lt;img src="google-search-for-the-best-2025-oled-monitors.png"
alt="Just a random Google search for the best 2025 OLED monitors"/>
&lt;/a>
&lt;/a>&lt;/span>&lt;span class="caption">&lt;span class="title">Just a random Google search for the best 2025 OLED monitors&lt;/span>&lt;/span>&lt;/span>&lt;/p>
&lt;p>There must be a better way! And thanks to Free and Open Source Software (FOSS), there is! &lt;a href="https://github.com/searxng/searxng"target="_blank" rel="noopener">SearXNG&lt;/a> is a free Internet metasearch engine that can be self-hosted and aggregates results from various search services such as Google, Bing, DuckDuckGo, Seznam, Qwant, etc. Using metasearch is like self-hosted search aggregation where users are not tracked or profiled and you get rid of ads as a bonus.&lt;/p>
&lt;p>I decided to give SearXNG a try because: &amp;ldquo;No tracking. No ads. Search like it used to be.&amp;rdquo; sounds just right. With some help from &lt;a href="https://docs.searxng.org/"target="_blank" rel="noopener">SearXNG Docs&lt;/a> I got my instance up and running. And at first glance, it is better. Much better! Just look at it:&lt;/p>
&lt;p>&lt;span class="image-container">&lt;span class="link">
&lt;a href="#searxng-search-for-the-best-2025-oled-monitors.png">
&lt;img class="img" src="searxng-search-for-the-best-2025-oled-monitors.png"
alt="Just a random SearXNG search for the best 2025 OLED monitors"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="searxng-search-for-the-best-2025-oled-monitors.png">
&lt;img src="searxng-search-for-the-best-2025-oled-monitors.png"
alt="Just a random SearXNG search for the best 2025 OLED monitors"/>
&lt;/a>
&lt;/a>&lt;/span>&lt;span class="caption">&lt;span class="title">Just a random SearXNG search for the best 2025 OLED monitors&lt;/span>&lt;/span>&lt;/span>&lt;/p>
&lt;p>Only relevant results. No ads. Multiple search engine results. Suggestions for phrase optimization and even Reddit community results on the first page. Incredible! The decision is made. I am migrating to SearXNG on my own instance as the default search on all my devices. But how do I replace default search engine in Mozilla Firefox and Google Chrome?&lt;/p>
&lt;h2 id="replace-the-default-search-engine-in-mozilla-firefox">Replace the default search engine in Mozilla Firefox&lt;/h2>
&lt;p>This is quite easy. Just follow the official &lt;a href="https://support.mozilla.org/en-US/kb/add-or-remove-search-engine-firefox"target="_blank" rel="noopener">Mozilla support page&lt;/a>. In short to add a search engine that supports &lt;a href="https://en.wikipedia.org/wiki/OpenSearch"target="_blank" rel="noopener">OpenSearch&lt;/a> from the address bar:&lt;/p>
&lt;ol>
&lt;li>Navigate to your SearXNG instance URL or just select one from the &lt;a href="https://searx.space/"target="_blank" rel="noopener">SearXNG public instances list&lt;/a>&lt;/li>
&lt;li>Select &lt;code>Add [Your SearXNG Instance Name]&lt;/code> option from the context menu&lt;/li>
&lt;li>Navigate to &lt;a href="about:preferences#search">about:preferences#search&lt;/a>&lt;/li>
&lt;li>Set your &lt;code>Default Search Engine&lt;/code> to the newly created entry in the list&lt;/li>
&lt;/ol>
&lt;h2 id="replace-the-default-search-engine-in-google-chrome">Replace the default search engine in Google Chrome&lt;/h2>
&lt;p>Google makes it quite tricky to find exactly where you should configure a custom search engine. This is not too surprising. As search Google Chrome and search is Google&amp;rsquo;s main interest. But it can be done:&lt;/p>
&lt;ol>
&lt;li>In Google Chrome navigate to &lt;a href="chrome://settings/searchEngines">chrome://settings/searchEngines&lt;/a>&lt;/li>
&lt;li>Leave the &lt;code>Keyboard shortcut&lt;/code> as it is&lt;/li>
&lt;li>Click &lt;code>Add&lt;/code> under &lt;code>Site search&lt;/code>&lt;/li>
&lt;li>Give your new search provider a &lt;code>Name&lt;/code> like &lt;code>SearXNG&lt;/code>&lt;/li>
&lt;li>For &lt;code>Shortcut&lt;/code> try something like &lt;code>@searxng&lt;/code>&lt;/li>
&lt;li>Fill in &lt;code>URL with %s in place of query&lt;/code> with &lt;a href="https://search.example.com/search?q=%25s"target="_blank" rel="noopener">https://search.example.com/search?q=%s&lt;/a> replacing the &lt;code>search.example.com&lt;/code> with your SearXNG instance&lt;/li>
&lt;li>Click &lt;code>Add&lt;/code> once the fields are complete&lt;/li>
&lt;li>Make this new &lt;code>Site search&lt;/code> provider your Chrome default by selecting the 3 dot menu beside the &lt;code>SearXNG&lt;/code> entry, selecting &lt;code>Make default&lt;/code>.&lt;/li>
&lt;/ol>
&lt;p>&lt;span class="image-container">&lt;span class="link">
&lt;a href="#replace-the-default-search-engine-in-google-chrome.png">
&lt;img class="img" src="replace-the-default-search-engine-in-google-chrome.png"
alt="Replace the default search engine in Google Chrome"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="replace-the-default-search-engine-in-google-chrome.png">
&lt;img src="replace-the-default-search-engine-in-google-chrome.png"
alt="Replace the default search engine in Google Chrome"/>
&lt;/a>
&lt;/a>&lt;/span>&lt;span class="caption">&lt;span class="title">Replace the default search engine in Google Chrome&lt;/span>&lt;/span>&lt;/span>&lt;/p>
&lt;p>Now when you type a search query in the Google Chrome address bar, your search will be routed through your SearXNG instance. Happy searching!&lt;/p></description></item><item><title>Zastav se a představ si. Jsi jiný.</title><link>https://use.cat/post/zastav-se-a-predstav-si-jsi-jiny/</link><pubDate>Thu, 26 Dec 2024 00:00:00 +0000</pubDate><guid>https://use.cat/post/zastav-se-a-predstav-si-jsi-jiny/</guid><description>&lt;p>Díky mojí moudré a milované sestře &lt;strong>Evě Veselé&lt;/strong>, že vdechla tomuto mému autorskému textu ženskou duši.&lt;/p>
&lt;h2 id="zastav-se-a-představ-si-jsi-jiný">Zastav se a představ si. Jsi jiný.&lt;/h2>
&lt;p>Máš jen jednu ruku. Asi abys mohl ještě cítit dotyk a hladit. Řídit koně, ale už nemohl nikdy tolik ublížit, když tě ovládne hněv.&lt;/p>
&lt;p>Máš jen jednu nohu. Asi abys věděl, že tvoje místo je tady a teď. Že nemusíš stále někam utíkat ani kráčet. Stačí stát. Znát směr a vědět, že bez svých blízkých daleko nedojdeš.&lt;/p>
&lt;p>Máš jen jedno oko. Asi proto, abys stále viděl a zřel krásy světa. Ale aby ti byl ukrytý prostor. Ten prostor, který nemůžeš a ani nemusíš ovládat.&lt;/p>
&lt;p>Máš jen jedno ucho. Asi abys musel víc napínat sluch a být vždy blíže tomu, kdo ti chce něco sdělit. Tak abys méně poslouchal a více naslouchal. Abys slyšel méně, ale hlouběji. Ty šepotavé vánky hlasů vnitřních.&lt;/p>
&lt;p>Zastav se a představ si. Jsi jiný.&lt;/p>
&lt;p>Máš dva nosy. Asi abys mohl lépe cítit. Cítit pachy smutku a strachu, vůně lásky, vzrušení, spadaného listí, čerstvé hlíny a kouře z ohně.&lt;/p>
&lt;p>Máš dvoje ústa. Asi abys ne více mluvil, ale více líbal. Asi abys mohl všem, na kterých ti záleží, častěji říct, jak moc je miluješ. A jak moc jim děkuješ za to, že jsou.&lt;/p>
&lt;p>Máš dvě srdce. Asi abys mohl lépe vnímat svůj tep, harmonii bytí a toho srdce bití. Sledovat svůj vnitřní rytmus. Abys mohl milovat všechny ty, kteří si to zaslouží, kteří po lásce touží, přestože i v jednom srdci je vždy dost místa pro každého čisté duše a dobré vůle.&lt;/p>
&lt;p>Máš dva mozky. Asi abys konečně pochopil, že života ti stále ubývá. No a jeho zbytek využil k tomu, abys pochopil, uchopil a nezapomněl všechnu tu krásu daru života.&lt;/p>
&lt;p>V součtu máš všeho stejně. Jenom jsi jiný. Hodně jiný. Představ si a zastav se. Zastav se a představ si.&lt;/p>
&lt;p>Jsi snílek? Nikoliv. Konečně si vážíš toho, čím jsi. Co máš. Sám v sobě. Kolem sebe.&lt;/p>
&lt;p>Rozhlédni se. Jsi jiný.&lt;/p>
&lt;p>&lt;span class="image-container">&lt;span class="link">
&lt;a href="#jane-goodall-05-hugo-van-national-geographic.jpg">
&lt;img class="img" src="jane-goodall-05-hugo-van-national-geographic.jpg"
alt="Jane Goodall 05, Photography by Hugo Van Lawick / National Geographic Creative"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="jane-goodall-05-hugo-van-national-geographic.jpg">
&lt;img src="jane-goodall-05-hugo-van-national-geographic.jpg"
alt="Jane Goodall 05, Photography by Hugo Van Lawick / National Geographic Creative"/>
&lt;/a>
&lt;/a>&lt;/span>&lt;span class="caption">&lt;span class="title">Jane Goodall 05, Photography by Hugo Van Lawick / National Geographic Creative&lt;/span>&lt;/span>&lt;/span>&lt;/p></description></item><item><title>Resume</title><link>https://use.cat/resume/</link><pubDate>Fri, 21 Jun 2024 00:00:00 +0000</pubDate><guid>https://use.cat/resume/</guid><description>&lt;!--
> **&amp;#57938; NOTE:** This is a rolling update article, where the publish date means the last update date. New skills are added and/or updated as I acquire, achieve or learn them.
-->
&lt;!--
URL PARAMS:
?hide=soft // hides soft achievements
?hide=hard // hides hard achievements
&amp;jobs=2 // hides jobs with prio 2 &amp; 3
&amp;jobs=3 // hides jobs with prio 3
eg. https://use.cat/resume/?hide=soft&amp;jobs=3 // hides soft achievements and jobs with prio 3
-->
&lt;h2 id="václav-veselý">Václav VESELÝ&lt;/h2>
&lt;p>Born &lt;a href="https://en.wikipedia.org/wiki/1983"target="_blank" rel="noopener">1983&lt;/a> in &lt;a href="https://en.wikipedia.org/wiki/Prague"target="_blank" rel="noopener">Prague, Czechia&lt;/a>. Czech &lt;sup>🇨🇿&lt;/sup> native, speaks English &lt;sup>🇺🇸&lt;/sup> and Spanish &lt;sup>🇪🇸&lt;/sup>. On the 2002 successfully finished leaving exam at the &lt;a href="https://en.wikipedia.org/wiki/Lyceum"target="_blank" rel="noopener">lyceum&lt;/a>. &lt;strong>2011 graduated&lt;/strong> from the &lt;a href="https://unicornuniversity.net/en/"target="_blank" rel="noopener">Unicorn University&lt;/a>. Let&amp;rsquo;s connect on &lt;a href="https://www.linkedin.com/in/vaclavvesely/"target="_blank" rel="noopener">LinkedIn&lt;/a>. Follow the &lt;a href="https://potla.ch/@vaclav"target="_blank" rel="noopener">Mastodon 🐘 profile&lt;/a>. Exchange an &lt;a href="mailto:vesely.vaclav&amp;#43;resume-antispam@gmail.com">email&lt;/a>. &lt;span id="phone">Place a call &lt;a href="tel:&amp;#43;420724286737">+420 724 286 737&lt;/a>.&lt;/span>&lt;/p>
&lt;h3 id="professional-summary">Professional summary&lt;/h3>
&lt;p class="profile">
&lt;a href="#/assets/img/vaclav-vesely-profile-photo-2017-07-14-color.jpg">
&lt;img src="https://use.cat/assets/img/vaclav-vesely-profile-photo-2017-07-14-color.jpg"
alt="Profile photo"/>
&lt;/a>
&lt;a href="#!" class="overlay" id="/assets/img/vaclav-vesely-profile-photo-2017-07-14-color.jpg">
&lt;img src="https://use.cat/assets/img/vaclav-vesely-profile-photo-2017-07-14-color.jpg"
alt="Profile photo"/>
&lt;/a>
&lt;/a>&lt;/p>
&lt;p>&lt;strong>Over 21 years&amp;rsquo; experience&lt;/strong> in &lt;a href="https://en.wikipedia.org/wiki/Information_and_communications_technology"target="_blank" rel="noopener">#ICT&lt;/a> analysis, architecture, sales, delivery and support. A balanced mix of &lt;span class="soft">&lt;strong>Technical Customer Success Manager&lt;/strong>, &lt;strong>Sales Engineer&lt;/strong>, &lt;strong>Systems &amp;amp; Business Analyst&lt;/strong>. &lt;/span>&lt;span class="hard">&lt;strong>Infrastructure Engineer&lt;/strong>, &lt;strong>DevOps&lt;/strong>, &lt;strong>DevSecOps&lt;/strong>. &lt;/span>Goal-oriented behaviours and networks of a &lt;span class="soft">&lt;strong>Head of Technical Sales, Customer Success Lead&lt;/strong> &lt;/span>&lt;strong>&lt;a href="https://en.wikipedia.org/wiki/Chief_information_officer"target="_blank" rel="noopener">CIO&lt;/a>&lt;/strong> or &lt;strong>&lt;a href="https://en.wikipedia.org/wiki/Chief_technology_officer"target="_blank" rel="noopener">CTO&lt;/a>&lt;/strong>. Have worked in &lt;strong>start-ups with 5–10 people&lt;/strong> &amp;amp; &lt;strong>large international corporations&lt;/strong>. So far &lt;strong>have directly led smaller teams of ~5 people for more than 5 years&lt;/strong> with excellent results. Have worked in &lt;strong>Technical Sales&lt;/strong> using &lt;a href="https://en.wikipedia.org/wiki/Qualified_prospect"target="_blank" rel="noopener">#BANT&lt;/a>, &lt;a href="https://en.wikipedia.org/wiki/Solution_selling"target="_blank" rel="noopener">#SPIN&lt;/a>, &lt;a href="https://en.wikipedia.org/wiki/The_Challenger_Sale"target="_blank" rel="noopener">Challenger Sales&lt;/a>, &lt;a href="https://meddicc.com/"target="_blank" rel="noopener">#MEDDIC&lt;/a> and many other &lt;strong>Solution Sales&lt;/strong> techniques.&lt;/p>
&lt;p>Have used so far &lt;a href="https://cloud.google.com/"target="_blank" rel="noopener">Google Cloud&lt;/a>, &lt;a href="https://aws.amazon.com/"target="_blank" rel="noopener">Amazon AWS&lt;/a>, &lt;a href="https://www.salesforce.com/products/platform/overview/"target="_blank" rel="noopener">Salesforce Platform&lt;/a> but prefer self-managed &lt;a href="https://en.wikipedia.org/wiki/Platform_as_a_service"target="_blank" rel="noopener">PaaS&lt;/a>. Have &lt;strong>Linux on all machines for 10+ years&lt;/strong>. Understand software development &amp;amp; delivery via #Agile &lt;a href="https://en.wikipedia.org/wiki/Scrum_%28software_development%29"target="_blank" rel="noopener">#SCRUM&lt;/a>. Personally code &amp;amp; script with &lt;strong>Shell script #Bash&lt;/strong>, &lt;strong>Python&lt;/strong>, &lt;strong>JavaScript&lt;/strong>. Prefer &lt;strong>infrastructure as a code&lt;/strong> &lt;a href="https://en.wikipedia.org/wiki/Infrastructure_as_code"target="_blank" rel="noopener">#IaC&lt;/a>. Deliver &lt;strong>source-code reproducible, Docker based&lt;/strong> build environments &amp;amp; containerized applications. &lt;span class="hard">Host own &lt;a href="https://git-scm.com/"target="_blank" rel="noopener">#Git&lt;/a> version control server based on &lt;a href="https://gitea.io/"target="_blank" rel="noopener">Gitea&lt;/a>. Use self-hosted &lt;a href="https://www.drone.io/"target="_blank" rel="noopener">Drone CI&lt;/a> &amp;amp; &lt;a href="https://www.jenkins.io/"target="_blank" rel="noopener">Jenkins&lt;/a> for &lt;strong>&lt;a href="https://en.wikipedia.org/wiki/CI/CD"target="_blank" rel="noopener">#CI/CD&lt;/a>&lt;/strong>.&lt;/span>&lt;/p>
&lt;p>Use web services, microservices, &lt;strong>CI/CD&lt;/strong> pipelines &amp;amp; &lt;strong>open-source software and hardware where possible&lt;/strong>. Mostly based on &lt;a href="https://www.debian.org/"target="_blank" rel="noopener">Debian&lt;/a> &lt;strong>Linux&lt;/strong> distributions, with &lt;strong>&lt;a href="https://www.docker.com/"target="_blank" rel="noopener">#Docker&lt;/a> containers&lt;/strong>, &lt;a href="https://docs.docker.com/compose/"target="_blank" rel="noopener">docker-compose&lt;/a> or &lt;a href="https://kubernetes.io/"target="_blank" rel="noopener">Kubernetes&lt;/a> with some &lt;a href="https://docs.gitlab.com/ee/ci/"target="_blank" rel="noopener">GitLab CI/CD&lt;/a> on top. &lt;span class="hard">Personally manage &lt;strong>cloud deployments&lt;/strong> &amp;amp; self-managed local &lt;strong>AMD, ARM or RISC-V&lt;/strong> based servers, routers &amp;amp; &lt;a href="https://en.wikipedia.org/wiki/Internet_of_things"target="_blank" rel="noopener">IoT&lt;/a>. Daily use &lt;a href="https://en.wikipedia.org/wiki/Terminal_emulator"target="_blank" rel="noopener">Linux terminal&lt;/a> mostly via &lt;a href="https://cs.wikipedia.org/wiki/Secure_Shell"target="_blank" rel="noopener">secure shell / #SSH&lt;/a> or even &lt;a href="https://en.wikipedia.org/wiki/Universal_asynchronous_receiver-transmitter"target="_blank" rel="noopener">UART&lt;/a>.&lt;/span> Expert on &lt;a href="https://www.google.com/sheets/about/"target="_blank" rel="noopener">Google Sheets&lt;/a>, including many of the &lt;a href="https://developers.google.com/apis-explorer"target="_blank" rel="noopener">Google APIs&lt;/a>, &lt;a href="https://developers.google.com/apps-script"target="_blank" rel="noopener">Google Apps Script&lt;/a>. Can do SQL / NoSQL databases queries like &lt;a href="https://www.postgresql.org/"target="_blank" rel="noopener">PostgreSQL&lt;/a> / &lt;a href="https://www.mongodb.com/"target="_blank" rel="noopener">MongoDB&lt;/a>. &lt;strong>Knows BigData tools&lt;/strong> like &lt;a href="https://pandas.pydata.org/"target="_blank" rel="noopener">Pandas&lt;/a> an &lt;a href="https://pola.rs/"target="_blank" rel="noopener">Polars&lt;/a>.&lt;/p>
&lt;p>Proficient in networking, &lt;a href="https://en.wikipedia.org/wiki/Domain_Name_System"target="_blank" rel="noopener">DNS&lt;/a>, &lt;a href="https://traefik.io/"target="_blank" rel="noopener">Traefik&lt;/a> &lt;strong>reverse proxy &amp;amp; load balancing&lt;/strong>, &lt;a href="https://letsencrypt.org/"target="_blank" rel="noopener">Let&amp;rsquo;s Encrypt&lt;/a>, &lt;a href="https://www.nginx.com/"target="_blank" rel="noopener">NGINX&lt;/a> and &lt;a href="https://httpd.apache.org/"target="_blank" rel="noopener">Apache HTTP Server&lt;/a>. &lt;span class="hard">Networking experience involves &lt;strong>Linux-based routers and firewalls&lt;/strong> like &lt;a href="https://mikrotik.com/"target="_blank" rel="noopener">Mikrotik RouterOS&lt;/a>, &lt;a href="https://openwrt.org/"target="_blank" rel="noopener">OpenWRT&lt;/a>, &lt;a href="https://www.pfsense.org/"target="_blank" rel="noopener">pfSense&lt;/a> and Debian or &lt;a href="https://www.alpinelinux.org/"target="_blank" rel="noopener">Alpine&lt;/a> based distributions as routers / WiFi access points.&lt;/span> Maintain several &lt;a href="https://www.wireguard.com/"target="_blank" rel="noopener">Wireguard&lt;/a>, &lt;a href="https://openvpn.net/"target="_blank" rel="noopener">OpenVPN&lt;/a> &amp;amp; &lt;a href="https://tailscale.com/"target="_blank" rel="noopener">Tailscale&lt;/a> meshes. Familiar with &lt;a href="https://cs.wikipedia.org/wiki/Representational_State_Transfer"target="_blank" rel="noopener">REST APIs&lt;/a>.&lt;span class="hard"> &lt;strong>Self manage services&lt;/strong> like this &lt;a href="https://use.cat/"target="_blank" rel="noopener">blog&lt;/a>, &lt;a href="https://ictoi.com/"target="_blank" rel="noopener">web of my business ICTOI&lt;/a>, &lt;a href="https://joinmastodon.org/"target="_blank" rel="noopener">Mastodon&lt;/a> instance &lt;a href="https://potla.ch/"target="_blank" rel="noopener">Potla.ch&lt;/a>, &lt;a href="https://rss.vesely.us/"target="_blank" rel="noopener">RSS reader&lt;/a>, &lt;a href="https://ntfy.sh/"target="_blank" rel="noopener">NTFY&lt;/a>, &lt;a href="https://photoprism.app/"target="_blank" rel="noopener">PhotoPrism&lt;/a>, &lt;a href="https://plausible.io/"target="_blank" rel="noopener">Plausible&lt;/a>, &lt;a href="https://nodered.org/"target="_blank" rel="noopener">Node-RED&lt;/a>, &lt;a href="https://matrix.org/"target="_blank" rel="noopener">Matrix&lt;/a>, &lt;a href="https://kanboard.org/"target="_blank" rel="noopener">Kanboard&lt;/a>, &lt;a href="https://kimai.org/"target="_blank" rel="noopener">Kimai&lt;/a>, &lt;a href="https://guacamole.apache.org/"target="_blank" rel="noopener">Apache Guacamole&lt;/a>, &lt;a href="https://tube.use.cat/"target="_blank" rel="noopener">PeerTube&lt;/a> with high availability &lt;strong>#HA&lt;/strong>.&lt;/span>&lt;/p>
&lt;h3 id="professional-goals">Professional goals&lt;/h3>
&lt;p>Maintain high &lt;strong>technical knowledge&lt;/strong>. Make time for &lt;strong>deep work&lt;/strong> often. Use &lt;strong>open source&lt;/strong>. Support &lt;strong>communities&lt;/strong>. Be open but value privacy. Be around &lt;strong>inspiring people&lt;/strong>. &lt;strong>Learn new skills&lt;/strong> every day. Get &lt;strong>hired for attitude&lt;/strong>. Become a better person.&lt;/p>
&lt;h3 id="work-experience-and-achievements">Work experience and achievements&lt;/h3>
&lt;div class="1">
&lt;h4 id="appsatori">AppSatori&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://appsatori.eu/"target="_blank" rel="noopener">AppSatori&lt;/a>&lt;/strong>, Period: &lt;strong>June 2024 – present&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong>, Contract type: &lt;strong>Contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Head of Strategic Partnerships&lt;/strong> &lt;sup>#GoogleCloud #GoogleWorkspace #GCP&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="uni"
>Achieved &lt;strong>Google Cloud Didigtal Leader Certification&lt;/strong> available online on my &lt;a href="https://www.credly.com/badges/1d185b7b-0fb3-4e2a-96f5-4545448b963d/public_url"target="_blank" rel="noopener">Credly profile&lt;/a>&lt;/li>
&lt;li
class="uni"
>Achieved &lt;strong>multiple Google Skill Boost badges&lt;/strong> available online on my &lt;a href="https://partner.cloudskillsboost.google/public_profiles/497b1b12-e0fb-4698-b7e4-0a78175e7f75"target="_blank" rel="noopener">Google Cloud Skills Boost profile&lt;/a>.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="1">
&lt;h4 id="auditor--stöger--partner">Auditor / Stöger &amp;amp; Partner&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://auditor.eu/"target="_blank" rel="noopener">Auditor&lt;/a>&lt;/strong>, Period: &lt;strong>March 2023 – April 2024&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong> / &lt;strong>Horn, Austria &lt;sup>🇦🇹&lt;/sup>&lt;/strong>, Contract type: &lt;strong>Project&lt;/strong> based / &lt;strong>contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Internal ICT Audit &amp;amp; Digitalization&lt;/strong> | &lt;strong>Head of Digital&lt;/strong> &lt;sup>#ITAudit #ITOutsourcing #Digitalization&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="uni"
>&lt;strong>Reduced the cost&lt;/strong> of new on-prem servers &lt;strong>by almost 20%&lt;/strong>, while maintaining the same RTO, RPO and uptime, simply by optimising the required compute power, scope of delivery and redefining the utilisation range for the next 7 years of the server’s life.&lt;/li>
&lt;li
class="uni"
>Fully managed, provisioned continuous upgrade path, process ownerships and support scheme for self-hosted &lt;a href="https://nextcloud.com/"target="_blank" rel="noopener">NextCloud&lt;/a> instance with &lt;strong>more than 600 active users across 200+ active clients&lt;/strong> (as of 01/24) handling highly available and critical business data with more than 300 customer documents a day.&lt;/li>
&lt;li
class="hard"
>Designed, tendered and managed replacement of old projector &amp;amp; Jabra based &lt;strong>conference room set-up with 360° panoramic Meeting Owl cameras&lt;/strong> and high resolution screens for all 5 Czech meeting rooms.&lt;/li>
&lt;li
class="uni"
>Technically managed the migration from a legacy &lt;a href="https://en.wikipedia.org/wiki/MS-DOS"target="_blank" rel="noopener">MS-DOS&lt;/a> based payroll system to a new cloud-based &lt;a href="https://en.wikipedia.org/wiki/Software_as_a_service"target="_blank" rel="noopener">SaaS&lt;/a> payroll system, &lt;strong>preserving the data history for over 2 000 payroll accounts, across 100+ active clients&lt;/strong> (as of 01/24) while maintaining key client touch points.&lt;/li>
&lt;li
class="uni"
>Executed a company-wide analysis and implementation strategy to &lt;strong>migrate all corporate secrets to fully managed, self-hosted password manager&lt;/strong> to upgrade the organization security and privacy to the next level.&lt;/li>
&lt;li
class="uni"
>Technically managed the &lt;strong>renegotiation of the contract for outsourced ICT services&lt;/strong> for the next couple of years.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="3">
&lt;h4 id="mosaic">Mosaic&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://www.mosaic51.com/"target="_blank" rel="noopener">Mosaic&lt;/a>&lt;/strong>, Period: &lt;strong>November 2022 – January 2023&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong>, Contract type: &lt;strong>Project&lt;/strong> based / &lt;strong>contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Infrastructure Engineer&lt;/strong> | &lt;strong>Technical Customer Success Consultant&lt;/strong> | &lt;strong>Sales Engineer&lt;/strong> &lt;sup>#AI #ComputerVision #B2B&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="uni"
>Consolidated, rewrite and restructuralize more than 30 mostly Google Documents, Spreadsheets &amp;amp; Presentations &lt;strong>covering completely Mosaic technical documentation&lt;/strong>. Moving from &lt;a href="https://en.wikipedia.org/wiki/WYSIWYG"target="_blank" rel="noopener">WYSIWIG&lt;/a> document approach to &lt;strong>online, self-contained, CI/CD built, Git revision controlled, &lt;a href="https://en.wikipedia.org/wiki/Markdown"target="_blank" rel="noopener">Markdown&lt;/a> based&lt;/strong> with rich media files and interactive 360° panoramas. Deployed online on a self-documented &lt;a href="https://en.wikipedia.org/wiki/Virtual_private_server"target="_blank" rel="noopener">VPS&lt;/a> &lt;a href="https://en.wikipedia.org/wiki/Infrastructure_as_code"target="_blank" rel="noopener">infrastructure as a code (IaC)&lt;/a> stack of &lt;strong>Debian Linux, Docker, Traefik, Drone CI, MkDocs, Material for MkDocs &amp;amp; NGINX&lt;/strong>. Creating a high valuable online asset for technical sales. Ready to serve almost &lt;strong>3 300 unique visitors a day&lt;/strong> (as of 01/23) at &lt;a href="https://docs.mosaic51.com/"target="_blank" rel="noopener">Mosaic Documentation&lt;/a>.&lt;/li>
&lt;li
class="uni"
>Streamlined the &lt;a href="https://www.hubspot.com/"target="_blank" rel="noopener">HubSpot&lt;/a> CRM process for lead capture via homepage form, &lt;a href="https://zapier.com/"target="_blank" rel="noopener">Zapier&lt;/a> to HubSpot integration for more automated and organized inbound lead processing.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="1">
&lt;h4 id="trustsoft-cloud">TrustSoft Cloud&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://www.trustsoft.eu/"target="_blank" rel="noopener">TrustSoft&lt;/a>&lt;/strong>, Period: &lt;strong>December 2021 – September 2022&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong>, mostly remote role, Contract type: &lt;strong>Project&lt;/strong> based / &lt;strong>contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Technical Customer Success Consultant&lt;/strong> | &lt;strong>Senior Consultant &amp;amp; Analyst&lt;/strong> | &lt;strong>Project &amp;amp; People Manager&lt;/strong> &lt;sup>#AWS #Cloud #B2B&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="uni"
>Managed a team of &lt;strong>5 DevOps Engineers&lt;/strong> to deliver a scale up project of &lt;a href="https://jira.atlassian.com/"target="_blank" rel="noopener">Atlassian Jira&lt;/a> service used by approximately &lt;strong>30 000 users&lt;/strong> in a world largest pharma company from single node on Amazon EC2 to a multi-node / high-availability architecture using &lt;strong>&lt;a href="https://aws.amazon.com/"target="_blank" rel="noopener">Amazon AWS&lt;/a> (ALB, AMI, ASG, EFS, RDS, S3, SG-ALB, SG-EC2, SG-RDS, VPC, Aurora), &lt;a href="https://www.sonarsource.com/products/sonarqube/"target="_blank" rel="noopener">SonarQube&lt;/a>, &lt;a href="https://jfrog.com/artifactory/"target="_blank" rel="noopener">JFrog Artifactory&lt;/a>, Atlassian Jira, Confluence &amp;amp; Crowd, &lt;a href="https://www.terraform.io/"target="_blank" rel="noopener">HashiCorp Terraform&lt;/a>, Gruntwork.io Terragrunt, Jenkins / CloudBees, Ansible &amp;amp; Vault, Docker &amp;amp; docker-compose, Git &amp;amp; BitBucket&lt;/strong>.&lt;/li>
&lt;li
class="uni"
>Reached goal of &lt;strong>Recovery Time Objective (RTO) of less than 8 hours&lt;/strong> and Recovery Point Objective (RPO) of 8 hours and deliver high availability, fail-over, performance increase, &lt;strong>instant scalability, rolling upgrades &amp;amp; disaster recovery&lt;/strong> by distributing the traffic between &lt;strong>multiple Jira nodes&lt;/strong>, upgrading to the latest &lt;a href="https://www.atlassian.com/"target="_blank" rel="noopener">Jira Enterprise&lt;/a> version, using Terraform for the infrastructure setup, switching to Aurora DB.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="1">
&lt;h4 id="simple-hardware">Simple Hardware&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://www.simplehw.eu/"target="_blank" rel="noopener">Simple Hardware&lt;/a>&lt;/strong>, Period: &lt;strong>February 2019 – September 2021&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong>, travels around the globe &lt;sup>🇯🇵 🇿🇦 🇹🇼 🇲🇽&lt;/sup>, Contract type: &lt;strong>Contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Head of IT / Chief Information Officer (CIO)&lt;/strong> | &lt;strong>Infrastructure Engineer&lt;/strong> | &lt;strong>Co-owner&lt;/strong> &lt;sup>#IoT #AI #ML #B2B2C&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="hard"
>Co-developed and managed a &lt;strong>parallelized hardware &amp;amp; software production line&lt;/strong> capable of &lt;strong>delivering hundreds of devices a day&lt;/strong> on &lt;strong>multiple product lines&lt;/strong> with central credentials management database &lt;strong>that produced more than 53 860 devices&lt;/strong> (as of 08/2021) using &lt;strong>Docker &amp;amp; docker-compose, Node-RED, Segger, UART, JTAG, REST API, PostgreSQL&lt;/strong>.&lt;/li>
&lt;li
class="hard"
>Deployed and managed product &amp;amp; services documentation and knowledge base on a stack of &lt;strong>Linux, Docker, Traefik, MkDocs, Material for MkDocs&lt;/strong>.&lt;/li>
&lt;li
class="hard"
>Created and managed to deploy and &lt;strong>run into production a highly structural, modular &amp;amp; versioned manufacturing documentation&lt;/strong> based on Markdown &amp;amp; &lt;a href="https://pandoc.org/"target="_blank" rel="noopener">Pandoc&lt;/a> to PDF generators for 4 products with tenths of product variants.&lt;/li>
&lt;li
class="soft"
>Co-managed, developed and executed the &lt;strong>growth of product portfolio from 1 to 8 products&lt;/strong> with &lt;strong>hundreds of product variants&lt;/strong>&lt;/li>
&lt;li
class="soft"
>Negotiated &lt;strong>#IoT&lt;/strong> device sales, project implementation &amp;amp; project rollout, presentations and public sales events, business development that helped to push opportunities to the later stages for the following clients: &lt;a href="https://www.generaliceska.cz/"target="_blank" rel="noopener">Generali Česká pojišťovna&lt;/a>, &lt;a href="https://www.direct.cz/"target="_blank" rel="noopener">Direct pojišťovna&lt;/a>, &lt;a href="https://www.kccs.co.jp/en/"target="_blank" rel="noopener">Kyocera Communication Systems (KCCS), Japan&lt;/a> and many others.&lt;/li>
&lt;li
class="soft"
>Directly sold and/or directly helped in a team to &lt;strong>increase the number of B2B partners from 0 to 419&lt;/strong> (as of 05/2020) in more than 60 countries around the globe in less than 2 years.&lt;/li>
&lt;li
class="soft"
>Prepared and updated &lt;strong>sales and marketing presentation&lt;/strong> that have been used in sales process for key clients and channel partners all over the globe.&lt;/li>
&lt;li
class="hard"
>Organized and personally processed on site &lt;strong>Japanese TELEC certification for 4 core radio products.&lt;/strong>&lt;/li>
&lt;li
class="uni"
>Co-managed and executed successful Simple Hardware &lt;strong>seed funding campaign via Fundlift&lt;/strong> platform for 6M CZK (~265k USD) &lt;strong>extended to 200%&lt;/strong> of the original target to &lt;strong>12M CZK (~530k USD)&lt;/strong> funded by &lt;strong>323 investors&lt;/strong>.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="1">
&lt;h4 id="simplecell-networks--heliot-europe">SimpleCell Networks / HELIOT Europe&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://www.sigfox.com/"target="_blank" rel="noopener">Sigfox&lt;/a>&lt;/strong>, Period: &lt;strong>February 2017 – February 2019&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong> / &lt;strong>Lausanne, Switzerland &lt;sup>🇨🇭&lt;/sup>&lt;/strong>, Contract type: &lt;strong>Contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Head of Sales (CSO)&lt;/strong> | &lt;strong>Technical Customer Success Consultant&lt;/strong> | &lt;strong>Sales Engineer&lt;/strong> &lt;sup>#IoT #B2B&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="hard"
>&lt;strong>Analyzed, implemented &amp;amp; migrated to new CRM&lt;/strong> and set up the complete pipeline lead to deal process and reporting.&lt;/li>
&lt;li
class="hard"
>&lt;strong>Completely streamlined the sales, quote to contract process&lt;/strong> &amp;amp; sales pipeline of Sigfox #IoT in Czechia.&lt;/li>
&lt;li
class="hard"
>&lt;strong>Co-organized the sales, quote to contract process&lt;/strong> &amp;amp; sales pipeline of Sigfox #IoT in &lt;strong>Heliot Switzerland&lt;/strong>.&lt;/li>
&lt;li
class="soft"
>&lt;strong>Achieved directly&lt;/strong> or with help of managed sales team (as of 11/2018): &lt;strong>32 356 activated Sigfox devices&lt;/strong>, &lt;strong>471 unique contracts / customers&lt;/strong>, &lt;strong>8.3 million CZK (~366k USD)&lt;/strong> committed volume on subscriptions.&lt;/li>
&lt;li
class="uni"
>&lt;strong>Co-organized and significantly contributed&lt;/strong> to the invitation process, marketing and technical sales on the &lt;a href="https://www.sigfox.com/en/sigfox-world-iot-expo-watch-replay"target="_blank" rel="noopener">Sigfox World IoT Expo 2017 in Prague on the 26th &amp;amp; 27th of September 2017&lt;/a>.&lt;/li>
&lt;li
class="soft"
>Negotiated subscriptions, implementation &amp;amp; change management contracts, presentations and public sales events, business development that helped to push opportunities to the later stages for the following clients: &lt;a href="https://www.generaliceska.cz/"target="_blank" rel="noopener">Generali Česká pojišťovna&lt;/a>, &lt;a href="https://www.direct.cz/"target="_blank" rel="noopener">Direct pojišťovna&lt;/a>, &lt;a href="https://www.direct.cz/"target="_blank" rel="noopener">SECURITAS Direct&lt;/a>, &lt;a href="https://www.direct.cz/"target="_blank" rel="noopener">EY&lt;/a>, &lt;a href="https://www.vekra.cz/"target="_blank" rel="noopener">VEKRA&lt;/a>, &lt;a href="https://www.stabilplastik.cz/"target="_blank" rel="noopener">Stabilplastik&lt;/a>, &lt;a href="https://www.innogy.cz/"target="_blank" rel="noopener">innogy Energie&lt;/a>, &lt;a href="https://www.t-mobile.cz/"target="_blank" rel="noopener">T-Mobile Czech&lt;/a>, &lt;a href="https://www.cez.cz/cs/o-cez/skupina-cez/vyznamne-spolecnosti-skupiny-cez/cez-ict-services"target="_blank" rel="noopener">ČEZ ICT Services&lt;/a>, &lt;a href="https://www.iot-water.cz/"target="_blank" rel="noopener">IoT.water&lt;/a>, &lt;a href="https://www.cdt.cz/"target="_blank" rel="noopener">ČD - Telematika&lt;/a>, &lt;a href="https://www.cleverfarm.ag/"target="_blank" rel="noopener">CleverFarm&lt;/a>, &lt;a href="https://www.cometsystem.cz/"target="_blank" rel="noopener">COMET SYSTEM&lt;/a>, &lt;a href="https://www.unis.cz/"target="_blank" rel="noopener">UNIS&lt;/a>, &lt;a href="https://www.appsatori.eu/"target="_blank" rel="noopener">Etnetera AppSatori&lt;/a>, &lt;a href="https://www.gordic.cz/"target="_blank" rel="noopener">GORDIC&lt;/a>, &lt;a href="https://www.aimtecglobal.com/"target="_blank" rel="noopener">AIMTEC&lt;/a>, &lt;a href="https://www.eman.cz/"target="_blank" rel="noopener">eMan&lt;/a>, &lt;a href="https://sensoneo.com/"target="_blank" rel="noopener">Sensoneo&lt;/a>, &lt;a href="https://www.protronix.cz/"target="_blank" rel="noopener">Protronix&lt;/a>, &lt;a href="https://www.elkoep.cz/"target="_blank" rel="noopener">ELKO EP&lt;/a>, &lt;a href="https://adastra.digital/en/"target="_blank" rel="noopener">ADASTRA&lt;/a> etc.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="1">
&lt;h4 id="pricewaterhousecoopers-pwc-polska">PricewaterhouseCoopers (PwC) Polska&lt;/h4>
&lt;p>Web: &lt;strong>&lt;a href="https://www.pwc.com/"target="_blank" rel="noopener">PwC&lt;/a>&lt;/strong>, Period: &lt;strong>November 2014 – January 2017&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong> / &lt;strong>Warsaw, Poland &lt;sup>🇵🇱&lt;/sup>&lt;/strong>, Contract type: &lt;strong>Contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>&lt;a href="https://www.salesforce.com/"target="_blank" rel="noopener">Salesforce.com&lt;/a> Technical Delivery Lead&lt;/strong> | &lt;strong>Sales Engineer&lt;/strong> | &lt;strong>Business Development Manager&lt;/strong> &lt;sup>#Salesforce.com #Cloud #B2B&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="soft"
>Reached turn-over point for the subsidiary in &lt;strong>48 months with 4 people on board and physical offices&lt;/strong>.&lt;/li>
&lt;li
class="uni"
>&lt;strong>Set up and managed the office for 5 people&lt;/strong> in &lt;a href="https://en.wikipedia.org/wiki/Dancing_House"target="_blank" rel="noopener">The Dancing House&lt;/a>.&lt;/li>
&lt;li
class="hard"
>&lt;strong>Hired core Czech Salesforce projects consulting &amp;amp; delivery team&lt;/strong> of &lt;a href="https://www.linkedin.com/in/martinhumpolec/"target="_blank" rel="noopener">Martin HUMPOLEC&lt;/a> &lt;a href="https://www.salesforce.com/blog/certified-technical-architects/"target="_blank" rel="noopener">Salesforce Certified Technical Architect (CTA)&lt;/a>, &lt;a href="https://www.linkedin.com/in/templehoff/"target="_blank" rel="noopener">Temple HOFF&lt;/a> &amp;amp; &lt;a href="https://www.linkedin.com/in/lukasrazim/"target="_blank" rel="noopener">Lukas RAZIM&lt;/a>.&lt;/li>
&lt;li
class="hard"
>Analysis, consulting, design and &lt;strong>development of the subsidiary landing page&lt;/strong> and event micro-pages using HTML, CSS, JavaScript, NodeJS, Heroku formerly on &lt;a href="http://www.outboxgroup.cz/"target="_blank" rel="noopener">outboxgroup.cz&lt;/a> then on &lt;a href="https://pwc-crm.cz/"target="_blank" rel="noopener">pwc-crm.cz&lt;/a>.&lt;/li>
&lt;li
class="soft"
>&lt;strong>Negotiated subscriptions, implementation &amp;amp; change management contracts&lt;/strong>, presentations and public sales events, business development that helped to push opportunities to the later stages in the sales pipeline for the following clients: &lt;a href="https://www.linet.com/en/"target="_blank" rel="noopener">LINET&lt;/a>, &lt;a href="https://www.equabank.cz/en"target="_blank" rel="noopener">Equa Bank&lt;/a>, &lt;a href="https://www.damejidlo.cz/en/"target="_blank" rel="noopener">DameJidlo.cz&lt;/a>, &lt;a href="https://www.psn.cz/"target="_blank" rel="noopener">PSN&lt;/a>, &lt;a href="https://www.ysoft.com/en"target="_blank" rel="noopener">YSoft&lt;/a>, &lt;a href="https://www.czc.cz/"target="_blank" rel="noopener">CZC&lt;/a>, &lt;a href="https://www.ittalents.cz/en/"target="_blank" rel="noopener">IT Talents&lt;/a>, &lt;a href="https://www.favorit.cz/"target="_blank" rel="noopener">Favorit&lt;/a>, &lt;a href="https://www.micos.cz/"target="_blank" rel="noopener">MICOS&lt;/a>, &lt;a href="https://www.zetor.cz/"target="_blank" rel="noopener">Zetor&lt;/a>, &lt;a href="https://cz.kruk.eu/"target="_blank" rel="noopener">KRUK&lt;/a>, &lt;a href="https://group.trenkwalder.com/en/"target="_blank" rel="noopener">Trenkwalder Group&lt;/a>, &lt;a href="https://www.izomat.cz/"target="_blank" rel="noopener">IZOMAT&lt;/a> etc.&lt;/li>
&lt;li
class="uni"
>&lt;strong>Signature and delivery of projects&lt;/strong> for the following clients: &lt;a href="https://www.edwards.com/gb/aboutus/global-locations"target="_blank" rel="noopener">Edwards Lifesciences Czech&lt;/a>, &lt;a href="https://www.vecr.cz/"target="_blank" rel="noopener">Veolia Energie ČR&lt;/a>, &lt;a href="https://www.muziker.cz/"target="_blank" rel="noopener">Muziker&lt;/a>, &lt;a href="https://www.ittalents.cz/en/"target="_blank" rel="noopener">IT Talents&lt;/a>, &lt;a href="https://www.tipsport.cz/"target="_blank" rel="noopener">TIPSPORT&lt;/a>, &lt;a href="https://ceskepreklady.cz/"target="_blank" rel="noopener">České překlady&lt;/a>, &lt;a href="https://www.fichtner.cz/"target="_blank" rel="noopener">Fichtner&lt;/a>, &lt;a href="https://www.ipex.cz/"target="_blank" rel="noopener">IPEX&lt;/a>, etc.&lt;/li>
&lt;/ul>
&lt;/div>
&lt;div class="2">
&lt;h4 id="netmail">Netmail&lt;/h4>
&lt;p>Web: &lt;strong>Netmail&lt;/strong> (now &lt;a href="https://www.revolgy.com/"target="_blank" rel="noopener">Revolgy&lt;/a>), Period: &lt;strong>May 2013 – December 2014&lt;/strong>, Places: &lt;strong>Prague, Czechia&lt;/strong>, Contract type: &lt;strong>Contractor&lt;/strong>, Roles &amp;amp; responsibilities: &lt;strong>Technical Customer Success Consultant&lt;/strong> | &lt;strong>Sales Engineer&lt;/strong> | &lt;strong>Google Apps Script Lecturer&lt;/strong> &lt;sup>#Google #Cloud #B2B&lt;/sup>&lt;/p>
&lt;ul>
&lt;li
class="uni"
>Introduced and conducted new &lt;strong>Google Apps Script&lt;/strong> Training service that &lt;strong>contributed 7 900 USD to revenue in less than 2 quarters&lt;/strong> of year 2014.&lt;/li>
&lt;li
class="soft"
>Prepared and maintained &lt;strong>see-think-do driven sales and marketing presentation&lt;/strong> templates that have been used in sales process for key clients and served as personalized offers by all sales representatives within Netmail.&lt;/li>
&lt;li
class="soft"
>Directly presented in more than &lt;strong>30 board member meetings / wide C-level audience presentations&lt;/strong> for the biggest Czech companies.&lt;/li>
&lt;li
class="soft"
>Directly sold and/or directly helped in a team that increased a number of &lt;strong>unique sold domains&lt;/strong> of Google Workspace from &lt;strong>140 on 5/2013 to 390 on 7/2014&lt;/strong> and a number of sold subscriptions &lt;strong>from 5 000 on the May 2013 to 20 300 on the July 2014&lt;/strong>.&lt;/li>
&lt;li
class="hard"
>&lt;strong>Trained more than 140 participants&lt;/strong> in &lt;strong>Google Apps Script, Google APIs&lt;/strong> and &lt;strong>Google Spreadsheets&lt;/strong> for top Czech companies like: &lt;a href="http://www.csas.cz/"target="_blank" rel="noopener">Česká spořitelna&lt;/a>, &lt;a href="http://www.veolia.cz/"target="_blank" rel="noopener">Veolia&lt;/a>, &lt;a href="http://www.copygeneral.cz/"target="_blank" rel="noopener">Copy General&lt;/a>, &lt;a href="http://www.economia.cz/"target="_blank" rel="noopener">Economia&lt;/a>, &lt;a href="https://www.btl.cz/"target="_blank" rel="noopener">BTL&lt;/a>, etc.&lt;/li>
&lt;li
class="soft"
>&lt;strong>Negotiated subscriptions, implementation &amp;amp; change management contracts&lt;/strong>, product and application development that helped to acquire or deliver products and services to the following clients: &lt;a href="http://www.csas.cz/"target="_blank" rel="noopener">Česká spořitelna&lt;/a>, &lt;a href="https://www.kofola.cz/"target="_blank" rel="noopener">Kofola&lt;/a>, &lt;a href="https://www.czc.cz/"target="_blank" rel="noopener">CZC&lt;/a>, &lt;a href="https://www.h1.cz/en"target="_blank" rel="noopener">H1&lt;/a>, &lt;a href="https://www.donauchem.com/"target="_blank" rel="noopener">Donauchem&lt;/a>, &lt;a href="https://www.annonce.cz/"target="_blank" rel="noopener">ANNONCE&lt;/a>, &lt;a href="https://www.theitalians.cz/"target="_blank" rel="noopener">Wine Food Market&lt;/a>, &lt;a href="https://www.fg.cz/"target="_blank" rel="noopener">FG Forrest&lt;/a>, &lt;a href="https://www.centrum.cz/"target="_blank" rel="noopener">Centrum Holdings&lt;/a>, etc.&lt;/li>
&lt;/ul>
&lt;/div></description></item></channel></rss>