Five months into writing about AI tooling on Hashnode, I noticed a pattern in the questions I get.

They're not about Claude or GPT. They're about CLAUDE.md.

Specifically: why people write a CLAUDE.md file, follow online guides, and still get the same frustrating result — Claude ignoring half of what they wrote, repeating the same mistakes, requiring constant re-prompting.

So I started AI Architecture Weekly — a free newsletter, every Monday, for senior devs who use AI daily and want depth instead of hype.

The first issue, out today, is about the five CLAUDE.md mistakes I see most often when I review production codebases:

1. Treating CLAUDE.md as a prompt
2. Listing rules without context
3. Skipping the project map
4. Generic patterns instead of project-specific ones
5. No "things that broke us" section

Each one is something that cost me hours before I learned to spot it. The full issue has concrete code examples and the fix I use now.

👉 Read the first issue: https://alexrogov.substack.com/p/5-claudemd-mistakes-that-quietly

The newsletter is free. It'll stay that way for the weekly issue. A paid tier opens in the fall with deeper case studies and full archive access — but the main weekly piece is always free.

If you've followed my Hashnode posts, the newsletter is the same voice with longer-form, more cohesive treatment of single topics. Less aggregator, more focused thinking.

What you'll get every Monday at 09:00 CET

• One specific pattern, mistake, or production story
• 1500–2000 words, no fluff
• Real code examples from real projects
• An opinion you can argue with

What it's NOT:

• Tool roundups
• Junior tutorials
• AI hype pieces

If a tip can't survive a year of refactors, you won't read it.

Subscribe

👉 https://alexrogov.substack.com

I'd love to hear what you're working on with AI tooling. Reply to the newsletter or comment here — your real problems shape what I write next.

— Alex

---

Originally published on my Hashnode blog. Follow me for more AI + Architecture content.