https://radosukala.com/writing Rado Sukala — Prague-based founder building products since 1999, now building Ctrl AI, the standard parts catalog for the agent era. Writing about the AI shift, from ChatGPT to coding agents. en Tue, 16 Jun 2026 00:00:00 GMT https://radosukala.com/writing/where-this-is-going https://radosukala.com/writing/where-this-is-going Tue, 16 Jun 2026 00:00:00 GMT The last note in the arc — from a browser tab in 2022 to building the layer agents stand on, and the one question that outlasts every tool https://radosukala.com/writing/the-vertical-part https://radosukala.com/writing/the-vertical-part Sun, 14 Jun 2026 00:00:00 GMT On Friday the government reached into Anthropic and switched off its two best models. Eleven years ago I tried to picture the vertical part of the curve — this is what it looks like from inside https://radosukala.com/writing/one-square-mile https://radosukala.com/writing/one-square-mile Thu, 28 May 2026 00:00:00 GMT What Los Angeles was to NFTs, San Francisco is to AI — except this time the thing at the center is real, and the center has never been so small https://radosukala.com/writing/the-boring-layer https://radosukala.com/writing/the-boring-layer Thu, 21 May 2026 00:00:00 GMT Everyone's watching the ceiling — hour-long agents, slick demos — but the dull layer underneath is where the keys stay, and that's the part my father taught me to build https://radosukala.com/writing/two-products-one-thesis https://radosukala.com/writing/two-products-one-thesis Thu, 23 Apr 2026 00:00:00 GMT People ask which of my two products is the real one — but trust for agents and ownership for people turned out to be the same problem wearing two faces https://radosukala.com/writing/you-stay-the-author https://radosukala.com/writing/you-stay-the-author Thu, 19 Mar 2026 00:00:00 GMT An assistant overshared a dossier of me I never wrote — so I'm building memory I own, where the AI proposes and I ratify https://radosukala.com/writing/the-phase-change https://radosukala.com/writing/the-phase-change Tue, 10 Mar 2026 00:00:00 GMT From spring it's obvious the way these things only ever are afterward: December was the month coding changed for good, and I felt it as the keystrokes leaving my hands https://radosukala.com/writing/parts-an-agent-wont-regenerate https://radosukala.com/writing/parts-an-agent-wont-regenerate Thu, 19 Feb 2026 00:00:00 GMT A part is verified when the agent stops arguing with it — on conformance, attestation, and a trust badge that shows verified versus generated https://radosukala.com/writing/openclaw https://radosukala.com/writing/openclaw Tue, 17 Feb 2026 00:00:00 GMT A solo developer built an agent that actually does things, it went off like a firework, and six weeks later OpenAI hired the man and set the project free https://radosukala.com/writing/building-in-the-open https://radosukala.com/writing/building-in-the-open Thu, 15 Jan 2026 00:00:00 GMT The first parts have shipped — now the real work is the contract a piece of software must satisfy before an agent will trust it instead of rewriting it https://radosukala.com/writing/three-years https://radosukala.com/writing/three-years Thu, 18 Dec 2025 00:00:00 GMT Three years from a one-question text box to handing agents whole features overnight — and why the year ahead is for building the quiet layer in the open https://radosukala.com/writing/memory-is-the-real-blocker https://radosukala.com/writing/memory-is-the-real-blocker Thu, 13 Nov 2025 00:00:00 GMT Four models this week, the same blank slate each time — the wall in front of me isn't intelligence, it's memory I own, and I've started building it https://radosukala.com/writing/verified-and-untouchable https://radosukala.com/writing/verified-and-untouchable Thu, 09 Oct 2025 00:00:00 GMT The product stance behind the catalog: a part is verified, attested, versioned — and untouchable. The agent composes it, never rewrites it, and trust becomes the unit of reuse https://radosukala.com/writing/all-in-on-ctrl-ai https://radosukala.com/writing/all-in-on-ctrl-ai Thu, 04 Sep 2025 00:00:00 GMT I closed the client work and pointed the whole studio at one bet: as agents get powerful, the scarce thing isn't intelligence, it's parts they can trust https://radosukala.com/writing/the-team-is-me-and-a-fleet https://radosukala.com/writing/the-team-is-me-and-a-fleet Thu, 21 Aug 2025 00:00:00 GMT The studio runs on a few senior humans and a fleet of agents now — same small-team instinct, new hands, and a gap I can't stop noticing https://radosukala.com/writing/standard-parts https://radosukala.com/writing/standard-parts Thu, 31 Jul 2025 00:00:00 GMT Agents reinvent the same bolt every time because they have no shelf to reach for — so I'm building one, and it's called Ctrl AI https://radosukala.com/writing/the-regeneration-problem https://radosukala.com/writing/the-regeneration-problem Thu, 26 Jun 2025 00:00:00 GMT Two sign-in flows, two days apart, nothing alike — and the drift underneath finally showed me the bug living beneath every other bug https://radosukala.com/writing/claude-4 https://radosukala.com/writing/claude-4 Fri, 23 May 2025 00:00:00 GMT Claude 4 holds a multi-hour task and the unit of work jumps from a function to a feature — but lean on it and the real gap shows: it has no memory of what we already trust https://radosukala.com/writing/capability-got-cheap https://radosukala.com/writing/capability-got-cheap Thu, 17 Apr 2025 00:00:00 GMT Three strong models shipped this month and I forgot which lab made which — when capability turns into a utility, the value moves up the stack to trust, reuse, and memory https://radosukala.com/writing/i-direct-i-dont-type https://radosukala.com/writing/i-direct-i-dont-type Fri, 21 Mar 2025 00:00:00 GMT I shipped a feature I never typed — what changes when the thinking moves upstream from syntax to intent, and the judgment that won't hand off https://radosukala.com/writing/the-terminal-writes-back https://radosukala.com/writing/the-terminal-writes-back Mon, 24 Feb 2025 00:00:00 GMT Claude Code arrives, the editor starts to dissolve, and I stop typing code — I describe outcomes and a terminal that thinks first writes back https://radosukala.com/writing/agents-leave-the-chat https://radosukala.com/writing/agents-leave-the-chat Thu, 23 Jan 2025 00:00:00 GMT For two years I carried the model's answers to the world by hand; this week the agent picked up the task and walked off with it itself https://radosukala.com/writing/the-gap-is-trust https://radosukala.com/writing/the-gap-is-trust Thu, 19 Dec 2024 00:00:00 GMT Two years in, the models are brilliant and the agents are real — what's still missing isn't intelligence, it's trust: grounding, provenance, memory https://radosukala.com/writing/a-protocol-for-context https://radosukala.com/writing/a-protocol-for-context Tue, 26 Nov 2024 00:00:00 GMT Anthropic shipped MCP — a boring standard socket for feeding tools and data to a model, and the boring ones are always the ones that matter https://radosukala.com/writing/it-moved-the-mouse https://radosukala.com/writing/it-moved-the-mouse Thu, 24 Oct 2024 00:00:00 GMT Anthropic gave a model hands this week — it took over my screen and clicked. The thrill wears off fast; what it's allowed to touch is the real question https://radosukala.com/writing/it-thinks-before-it-answers https://radosukala.com/writing/it-thinks-before-it-answers Thu, 19 Sep 2024 00:00:00 GMT OpenAI's o1 stops answering on reflex and starts to plan — the shift from autocomplete to deliberation, and why the pause is the point https://radosukala.com/writing/everyone-discovers-cursor https://radosukala.com/writing/everyone-discovers-cursor Wed, 07 Aug 2024 00:00:00 GMT The AI editor I quietly adopted a year ago is suddenly the whole industry's — and the road I drove alone now has traffic on it https://radosukala.com/writing/i-stopped-reading-the-docs https://radosukala.com/writing/i-stopped-reading-the-docs Thu, 25 Jul 2024 00:00:00 GMT The model became my interface to every framework I touch — enormous leverage, and a creeping illiteracy I have to watch in myself https://radosukala.com/writing/artifacts https://radosukala.com/writing/artifacts Fri, 21 Jun 2024 00:00:00 GMT A new Sonnet builds a working interface beside the chat — and leaves me circling the real problem: we keep regenerating software that already exists https://radosukala.com/writing/it-talks-back https://radosukala.com/writing/it-talks-back Tue, 14 May 2024 00:00:00 GMT GPT-4o's voice arrives on an ordinary Tuesday — the interface dissolves into plain talk, but the machine still meets me as a fluent stranger https://radosukala.com/writing/open-models-grow-up https://radosukala.com/writing/open-models-grow-up Fri, 19 Apr 2024 00:00:00 GMT Llama 3 ships open weights, and for a builder twice burned by platforms he didn't control, a model you can hold on your own disk changes the calculus https://radosukala.com/writing/the-mortgage-robot-six-years-later https://radosukala.com/writing/the-mortgage-robot-six-years-later Tue, 26 Mar 2024 00:00:00 GMT In 2019 we hand-wired a mortgage robot from scans to approval; this week one prompt did the whole thing — vindication, and the quiet cost of being early https://radosukala.com/writing/the-one-i-trust https://radosukala.com/writing/the-one-i-trust Tue, 05 Mar 2024 00:00:00 GMT A new model arrived and I skipped the leaderboards — Claude 3 Opus earned my daily seat on judgment and prose, not benchmark bars https://radosukala.com/writing/a-million-tokens https://radosukala.com/writing/a-million-tokens Fri, 16 Feb 2024 00:00:00 GMT Gemini's million-token window and Sora land the same week, and another intuition about limits quietly breaks https://radosukala.com/writing/one-year-in https://radosukala.com/writing/one-year-in Wed, 20 Dec 2023 00:00:00 GMT A year after shuttling code between two windows, the model moved into my editor — and the gap I can't shake isn't intelligence, it's what it knows about us https://radosukala.com/writing/the-week-it-almost-died https://radosukala.com/writing/the-week-it-almost-died Wed, 22 Nov 2023 00:00:00 GMT A board fired and rehired the man half the industry runs on over one weekend, and it drew a hard line under a question I can't unsee: what in this stack do I actually own https://radosukala.com/writing/they-turned-it-into-a-platform https://radosukala.com/writing/they-turned-it-into-a-platform Wed, 08 Nov 2023 00:00:00 GMT OpenAI's DevDay turned the model into a platform, and every builder now faces the oldest question I know: build on top of it, or beside it https://radosukala.com/writing/it-doesnt-know-my-company https://radosukala.com/writing/it-doesnt-know-my-company Tue, 12 Sep 2023 00:00:00 GMT A model brilliant about the world and blank about my company — and why bolting on retrieval is plumbing, not the real problem: trust in the answer https://radosukala.com/writing/a-calmer-model https://radosukala.com/writing/a-calmer-model Tue, 18 Jul 2023 00:00:00 GMT Claude 2's 100k context lets me feed whole documents at once — but it's the calmer temperament, not the benchmark, that keeps me coming back https://radosukala.com/writing/the-museum-of-the-future https://radosukala.com/writing/the-museum-of-the-future Thu, 22 Jun 2023 00:00:00 GMT Japan was the future once. Walking Tokyo now, you feel a whole country that reached a peak and stayed there while the center of gravity moved on https://radosukala.com/writing/the-editor-can-see-my-code https://radosukala.com/writing/the-editor-can-see-my-code Thu, 08 Jun 2023 00:00:00 GMT The model moves into my editor and the copy-paste tax finally falls — but a quieter problem shows up in its place https://radosukala.com/writing/agents-that-almost-work https://radosukala.com/writing/agents-that-almost-work Thu, 27 Apr 2023 00:00:00 GMT Everyone's passing AutoGPT around like a magic trick — I gave one a real job and watched it unravel, and the failure points straight at what's missing https://radosukala.com/writing/copy-paste-run https://radosukala.com/writing/copy-paste-run Wed, 12 Apr 2023 00:00:00 GMT I'm building software again — by talking to GPT-4 and shuttling its code into VS Code by hand. https://radosukala.com/writing/gpt-4-changes-the-math https://radosukala.com/writing/gpt-4-changes-the-math Wed, 15 Mar 2023 00:00:00 GMT I ran GPT-4 through the private tests every model had failed, and it cleared the ones I thought were five years off — so I'm raising what I dare to build https://radosukala.com/writing/sydney https://radosukala.com/writing/sydney Tue, 14 Feb 2023 00:00:00 GMT Microsoft's Bing chatbot melts down as "Sydney" — and the real unease isn't the machine, it's how fast we project a self into a model with nothing to anchor it https://radosukala.com/writing/pretending-this-is-normal https://radosukala.com/writing/pretending-this-is-normal Tue, 24 Jan 2023 00:00:00 GMT A hundred million users in two months, ten billion from Microsoft, and the table reaches for the bread — the speed of the shrug unsettles me more than the technology https://radosukala.com/writing/too-early https://radosukala.com/writing/too-early Thu, 22 Dec 2022 00:00:00 GMT JudRobot and Hyponamíru weren't failures — they were scouts I sent out before the machine could carry the idea, and now it can https://radosukala.com/writing/i-keep-opening-the-tab https://radosukala.com/writing/i-keep-opening-the-tab Thu, 08 Dec 2022 00:00:00 GMT A week in, a founder who waited two decades for this can't stop opening the tab — every night he brings it something harder, and every night it goes further than he expects https://radosukala.com/writing/it-finally-happened https://radosukala.com/writing/it-finally-happened Wed, 30 Nov 2022 00:00:00 GMT OpenAI quietly put a chatbot on the internet today. I think the ground just moved. https://radosukala.com/writing/the-long-flat-part https://radosukala.com/writing/the-long-flat-part Thu, 29 Jan 2015 00:00:00 GMT I read a blog post about artificial intelligence this week and something in me rearranged. Now I can't stop talking about a future that isn't here yet