Fata uses spaced repetition to counter AI-induced skill rot in developers
Djoumé, a 20-year developer, built Fata — a spaced-repetition app for coding fundamentals (Rust, CSS, React, Python, TypeScript, architecture) — after noticing his own recall degrading from relying on AI agents.
Score breakdown
Fata directly targets a concrete side-effect of the AI coding shift — degraded recall of fundamentals — by applying genuine SRS scheduling (not gamification) to developer skill maintenance, with a no-signup browser entry point that lowers the barrier to trying it.
- 01Built by Djoumé, a developer with over 20 years of experience, in response to personal skill decay from AI-agent-driven coding
- 02Covers Rust, CSS, React, Python, TypeScript, and architecture
- 03Uses a real SRS (spaced-repetition system) scheduler, described as Anki-like, not streak-based
Djoumé, a developer with over 20 years of experience, built Fata after noticing that relying almost exclusively on AI agents for coding caused his recall of fundamental skills to degrade — things he previously knew cold, he was now asking the agent to confirm. Fata addresses this with short daily spaced-repetition sessions covering Rust, CSS, React, Python, TypeScript, and software architecture. The scheduling engine is described as a real SRS system, comparable to Anki, explicitly not streak-based gamification.
Course content began as hand-written material but is now predominantly AI-generated, with each course requiring approximately 3,000 LLM calls to produce.
On the technical side, Fata is an offline-first mobile app built with Capacitor, using RxDB for local storage that syncs to Firestore. Course content began as hand-written material but is now predominantly AI-generated, with each course requiring approximately 3,000 LLM calls to produce. To maintain quality, every code sample is put through compilation, linting, unit testing, and both AI and manual review before it ships. Courses are accessible in the browser at fata.app with no account required.
Key facts
- 01Built by Djoumé, a developer with over 20 years of experience, in response to personal skill decay from AI-agent-driven coding
- 02Covers Rust, CSS, React, Python, TypeScript, and architecture
- 03Uses a real SRS (spaced-repetition system) scheduler, described as Anki-like, not streak-based
- 04Offline-first mobile app built with Capacitor; syncs via RxDB to Firestore
- 05Each course requires approximately 3,000 LLM calls to generate
- 06Every code sample goes through compilation, linting, unit testing, and AI and manual review before shipping
- 07Courses can be tried in the browser at fata.app with no signup required
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 12, 2026 · 10:05 UTC. How this works →