Jan Červinka

Skills, custom prompts, training — switching costs compound fast. Even with 5.5 looking exciting I'm still on current Opus, the friction isn't worth it. No surprise both anthropic and openai invest heavily in enterprise adoption. anthropic.com/news/enterpris…

Claude is excellent at building websites, apps, and slides. But when I tried to make a beautiful diagram, it stumbled. I expected more. Tried Pencil, web-based editors, Mermaid. Nothing worked well. Then D2 saved the day. Not amazing, but good enough. This discourages people. You hit something simple that turns out to be unexpectedly tricky with AI, and you want to drop it. Pushing through is where the learning actually happens. That was Friday, a couple of hours before Claude Design launched. I was curious to see how much better it would do. After playing with it today, not much. As always, I fall back to just Claude Code in the terminal. That said, I see real potential in Claude Design. The structured Design System prep is a good primitive for AI collaboration. Anyway, what's your stack for beautiful diagrams with AI?

@kevinrose Yes! I used to ask - “remind me, where were we?” Not anymore.

thank the heavens for the claude code mini-summaries at the bottom of terminal windows, sooo many tabs, this helps

This. I think catering to builders and entrepreneurs is one of the biggest opportunities right now. Everyone's a builder now.

As I moved out of daily development into more managerial roles, I thought I'd be writing emails most of the time. Not code. Well, AI changed that quite a bit. I'm still not writing code — but Claude is. Multiple things at once. While I'm on Google Meet. Turns out my MacBook Air with 16 gigs of RAM wasn't built for that. Got the M5 Pro yesterday, and moving everything over became a pain. Then I realized I could enlist Claude to help. Switched on SSH on the old Mac, asked Claude to connect, map the landscape, and carry over what matters to me now. Got Claude Code on the new Mac talking to Claude Code on the old one. Surreal and fun. Next time, this is where I'll start — though I expect this will all be quite different when next time actually comes.

Thats actually exactly why I prefer just a perfected Claude Code setup over OpenClaw. I feel I control the proces closer and focus on it. Telegraming my agent never felt good enough. Though the magic of having an agent with personality is lost with claude code

Breaking a reasonably protected system used to require someone skilled and motivated enough to dig. Most systems were safe by friction alone. Claude Mythos (however capable it turns out to be) is a reminder that friction is disappearing fast. The 27-year-old OpenBSD bug needed 1,000 parallel agent runs — but it still found something that survived decades of human review. Next generation, it probably won't need the 1,000 runs. We started running the best publicly available models against our own systems continuously. Basically an automated pentest. Because the alternative is letting less skilled, less motivated people find the issues first.

Several of our senior people are shifting from managing teams to managing AI agents. Becoming key contributors again instead of just coordinators. That transition needs protected time. But projects without a clear owner keep creeping in. They sound simple — "just a quick setup," "someone will handle it" — so nobody properly owns them. They accumulate and steal focus from the work that actually matters. I decided to sit down with some of our key people to realign once AI initiatives landed on everyone's plate. Cut meetings. Split responsibilities. Made explicit who owns what and how much capacity they actually have. Everything is possible now — AI made it all feel two prompts away. But doing things well still takes time and focus. Picking what not to do is the hard part.

I can realistically run about 3 things in parallel with AI coding agents. More than that, and some get stuck because I can't guide them properly. The bottleneck isn't the agents. It's my ability to maintain mental models of what each one is doing and intervene at the right moments. Three workstreams and I'm sharp. Four or five and something sits idle for 20 minutes because I forgot it was waiting on a decision. Most AI content focuses on what the tools can do. What I keep hitting is how many things I can actually hold in my head at once. Not sure if 3 is everyone's number or just mine.

The scary thing is that AI tools turned everyone into half a developer. Non-devs installing random packages through Claude. The attack surface expanded to every employee with an AI coding tool.

David Kasper keeps praising Pencil. Figma just released their MCP for design manipulation. So I compared them. We had existing website and slide deck designs in Figma. I asked Opus to create a new page and a few slides from a markdown outline — through Figma MCP, Pencil MCP, and Pencil's built-in agents. Both MCPs could do it but deviated from our brand significantly. Pencil agents were on another level. Faster, followed conventions exactly, mostly good to go. If you're experimenting with AI-assisted design, give Pencil a try. Great job @tomkrcha.

I'm thinking a lot about the balance between exploration and execution with AI tools. It's easy to keep searching for the best workflows. The best tools. The best stack. It's fun. And it matters — you need to know what's out there. But it's also where you can lose months. Should we use Compound Engineering or Superpowers? Codex or Claude Code? Build an internal marketplace? Every week there's a new question. I catch myself doing this — exploring when I should be executing. And it's not just me. It applies to the whole team. At some point you need enough people on the same setup that they can actually help each other. Share tricks. Build on each other's work. Maybe something 10% better exists out there. But we lose more from everyone still messing around than from just picking one and committing. It's time to pick one and get good at it. How do you find the right balance between exploration and execution in times like these?

@danshipper @garrytan @tobi Can confirm. Can't imagine leading a company without rewiring my brain to actually understand what's possible now. The ramp up is fast with CEOs though — the agency these tools bring and the immediate feedback loop gets intoxicating. And watching others extremely inspirational.

in ai, a company is only going to go as far as their CEO goes you cannot delegate working with the latest tools—it's a critical part of building your intuition for how this new world works that's why @garrytan @tobi etc are going so hard right now. bullish

The economics of meeting bots is a bit broken. You pay per user, but some people have one meeting a month and some have a hundred. They all sit in the same meeting and need only one set of notes. Notion Meeting Notes and other embedded solutions are great but they usually don't handle Czech perfectly. Or they need you to remember to start the recording. We wanted everyone to have meeting notes available seamlessly — no friction, no thinking about it. Only that way can we ensure we have sufficient docs. And those docs are what AI agents need to have context and actually be useful. So we spent a couple of days building a custom solution on top of @Recallai and @ElevenLabs Scribe — best combo for Czech and English we've found. It auto-joins from Google Calendar, works across Meet, Zoom, Teams, Slack Huddles. Notes land in Notion automatically. One person built it. After a few months in production, the cost is under a dollar per person per month. A fraction of what commercial alternatives charge. Honestly, we didn't build it to save money. We built it because good meeting notes are infrastructure for everything else we're doing with AI.

This actually mirrors how we tend to build innovative things now. But is this just because the current state of AI adoption requires pirates to keep pushing? We are starting to call them alchemists…

new model for engineering team structure in 2026: 2 people only one pirate and one architect the pirate's job is to move as fast as possible to develop valuable, shipped product features by vibe coding. the architect's job is to turn the product surface discovered by the pirate into a reliable, structured machine—also by vibe coding, but at a slower, more well-reasoned pace. every product needs a pirate but most product's only need an architect once they some form of PMF, and in that case they usually don't need one full-time. architects can work across many codebases and solve interesting technical challenges. pirates go hard on a product that they own end-to-end.

We had three internal demos yesterday—products that will go live very soon. All are industry-leading concepts. Only one was built by an actual developer, and none took longer than a month to create. What they required were amazing people with years of industry experience and drive, armed with powerful tools. Crazy times, and amazing to watch the team build. Let's go 🚀

Claude can now auto-compact the conversation; Claude Code style. No more abrupt ends. Game-changing for me and by far my biggest Claude frustration solved 🎉 Thank you, @claudeai

@mymind Hey guys, I wanted to share some feedback with you (I have an account) but the contact button on the helpscout page does not seem to work. I tried iOS and Mac, Safari and Chrome,… Happy to help with debugging if needed.

Purple Remote Office 2025. 💜 For a few days we were often offline, but more connected than ever. Just like every year, it was the time when we brought together people who, most of the year, are scattered all over the world. youtu.be/vXwfZNX84m4

🛑 Stop Making AI Do Your Thinking 🧠 --- Your brain is getting lazy. I love AI. In our Purple Technology, it's quickly becoming our daily driver.. It writes my reports, gathers my research, and makes me faster than ever. But here's the problem: AI is making you dumber. Not less productive. Less wise. I still remember a situation when I couldn't recall basic project details in a conversation. I'd read the AI summary that morning. The information was there. Then it wasn't. I consumed it. Never internalized it. Your brain works like a muscle. Use it or lose it. When you let AI do all your thinking, your thinking muscles atrophy. The result? You become a passive consumer of information instead of an active processor. When calculators arrived, we lost some mental math skills. When GPS took over, we stopped reading maps. But those were specific tools replacing specific skills. We adapted fine because we didn't need those particular abilities as much anymore. This is different. Dangerously different. When you outsource your thinking to AI—your ability to process information, connect dots, build understanding—you're not losing a skill. You're losing your capacity to develop wisdom. To grow intelligence itself. We're building artificial intelligence while letting our human intelligence atrophy. The research backs this up. It's called "cognitive offloading," and unlike losing the ability to calculate square roots in your head, this rewires your brain for dependence on the very thing that makes you human: your ability to think, understand, and create meaning from complexity. Here's what I did about it. I started writing by hand again. Sounds old-fashioned. It works. Handwriting forces your brain to choose. You can't capture everything. So you process. Summarize. Connect ideas. Your brain becomes a curator, not a recorder. My system has four parts ✅ Morning handwriting: I start each day by handwriting my goals and gathering information I'll need. When I skip this because I'm in a hurry, I immediately notice the difference. ✅ Think first, AI second: Write your thoughts before asking AI anything. Ask better questions: Don't consume AI reports passively. Question them. ✅ Teach the AI: Correct it. Guide it. This reinforces your own understanding. ✅ End-of-day voice notes: Talk through your handwritten notes. Let AI capture and organize this. Best of both worlds. This feels like extra work. It is. That’s the point. Every connection I make manually strengthens my brain connections. Every moment I think without AI builds cognitive muscle. Don't abandon AI. That's not the goal. Just don't let it abandon your ability to think. The future belongs to people who can dance with AI. Use it for what it's good at. Keep what you're good at. Use artificial intelligence. Build your intelligence too. I'm thinking this article through with a pen right now. Later, I'll use AI to polish it. The thinking is mine. The execution is ours. That's the balance I want.