A family saga in six volumes, generated by AI, narrated in my cloned voice. How I built a multi-agent pipeline — with Architect, Writer, Reviewer, and Summarizer — that writes coherent, multi-volume stories from a napkin sketch to audiobook overnight.
Read more
Flaky tests are a slow leak — small enough to ignore, persistent enough to compound. I built a system that investigates them, fixes them, and verifies the fix autonomously — with six specialized AI agents working in parallel.
Read more
I work on multiple machines and re-configuring Claude Code every time was tedious. I built a simple solution: a Git repo as the single source of truth, using symlinks instead of file copies.
Read more
QA taught me that without feedback, software development is a risk. In work with AI agents, TDD has become even more important. See how I use tests as "guardrails" and why this approach saves hours of tedious debugging.
Read more
I always wanted a vocabulary app that doesn’t teach me about "green apples" or "Aunt Alice," but about system architecture, Code Review, and Business English. I created AI English PRO – a personalized PWA that teaches me exactly what I need for work.
Read more