Launching Aleph

Today I’m launching Aleph. Name a topic, say how much you already know, and it generates a structured path you can start working through on your phone the same minute. You can try it in your browser or read the source on GitHub.

Two months ago I wrote that I didn’t want a standalone app for this — that a Discord bot, SQLite, and static HTML over Tailscale got me 95% of the way there. I still think that was the right call at the time. But, I wanted to ship functionality I couldn't with the simple static HTML (i.e. in lesson AI agents, etc).

What’s live today:

• A generated path — units and lessons for whatever topic you name, each lesson a short Read passage followed by a Quick check. Lessons generate on demand as you reach them, so you’re never waiting on a whole path.
• An AI tutor you can talk to inside a lesson, and that can reshape the rest of your path when you ask it to.
• Flashcards with spaced repetition, suggested from lessons you’ve actually finished.
• Streaks, and deliberately nothing more in the way of gamification.

The interesting part, again, was the shipping process:

• Docs before code. Every phase gets a PRD and a TDD before anything is written, and docs/CONTEXT.md pins the vocabulary — path, unit, lesson, Read passage, Quick check — so prose, prompts, and schemas all use the same word for the same thing. This turned out to be the single highest-leverage artifact for working with agents.

The through-line is the same one from Habagou: spend the effort on making the planning, delegation, and verification loops short, and let shipping a real product stay the fixed point. The difference this time is how much of that leverage came from documents rather than code.

---

This post was drafted in collaboration with Claude. Attributing when AI is part of the brainstorming or writing process and tagging those posts w/ ai-assisted.