🏠 Cardiologist wins 3rd place at Anthropic's hackathon

GM! It’s Brett. Last week, I rebuilt our entire landing page from scratch and it looks better than the landing page we spent $100K on for an agency to build.

Thanks to Claude Code.

What this means for everyone is that the only thing blocking you from actually building your ideas is..... you.

The tools are there. If you don't know how to do something, all you have to do is ask the tools.

Most people using Claude Code hand everything off to the AI, let it loop, and wonder why the results are inconsistent. The problem isn't the model. It's that you stopped being in control.

Here's what actually works:

Clear manually, don't wait for auto-compaction. When context fills up, Claude drops your context and quietly loses critical instructions. Use /clear between tasks and reload only what's relevant.

Be explicit, not hopeful. Don't assume Claude will use the right library, follow the right pattern, or check the docs. Tell it explicitly. If you want it to look up BetterAuth docs before implementing auth, say that. The AI might have done it anyway, but now you're sure.

Keep your CLAUDE.md under 100 lines. Treat it like guardrails, not a manual. Only document what Claude keeps getting wrong. When it gets too long, Claude starts ignoring half of it.

Use the AskUserQuestion tool before building. Add this to your CLAUDE.md: "If you need my input, use the AskUserQuestion tool to ask." Start with something like "Interview me to build a spec before you write anything." All ambiguity gets resolved before a single line is written.

Stay in control, don't loop. There's a lot of hype around running Claude Code in automated loops with a big plan doc. Ignore it. The results aren't convincing. You get better output when you stay involved at every step.

Use Plan mode before touching anything complex. Hit Shift+Tab before refactoring more than 3 files. Claude maps the territory before moving. Approve or redirect before a single line gets written.

Read targeted lines, not full files. Read auth.ts:50-120 costs ~1,200 tokens. The whole file costs ~4,000. Small habit, massive savings over time.

End every session with a handoff doc. Ask Claude to summarize decisions made, files changed, and what's next. Your next session picks up instantly.

The developers winning with these tools aren't better prompters. They're better context architects.

Watch the full video here.

Founders spend too much time drafting the same kinds of messages. Wispr Flow turns spoken thinking into final-draft writing so you can record investor updates, product briefs, and run-of-the-mill status notes by voice. Use saved snippets for recurring intros, insert calendar links by voice, and keep comms consistent across the team. It preserves your tone, fixes punctuation, and formats lists so you send confident messages fast. Works on Mac, Windows, and iPhone. Try Wispr Flow for founders.

Start flowing free

Michał Nedoszytko is a practicing cardiologist. Not a programmer. He built PostVisit.ai, an AI-powered patient care companion, in 7 days. Coded in hospitals, airports, and hotel rooms flying between Brussels and San Francisco.

Out of 13,000 submissions at Anthropic's hackathon, he placed 3rd.

A few years ago, this was simply not possible without an engineering team and months of runway.

Here's the real story though: we're entering an era where domain experts with an engineering mindset will outbuild traditional developers. A cardiologist understands patient drop-off after consultations better than any product manager ever could.

AI just removed the only barrier standing between that insight and a working product.

The most dangerous builders going forward won't be the best coders, they'll be the smartest domain experts.

Last week’s poll: If you guessed AI-generated… you were right.

🟨🟨🟨⬜️⬜️⬜️ Yeah (~48%)

🟩🟩🟩⬜️⬜️⬜️ Nope (~52%)

Here’s the original.

Micro is a CRM that doesn’t feel like homework and an email client that knows what matters. Simply put, it’s an all-in-one productivity tool that organizes itself so you don’t have to.

LHV is a community-driven, early-stage venture fund backing ambitious founders at the earliest stages.

If you’re building tools that help founders, investors and other professionals who want to be 100x more productive with 100x less work, and you’re interested in advertising with us, send an email over to [email protected] with the subject “Homescreen Ads”