Leonid M's articles
- about 2 months agoAI Coding
Four Ways Agent Context Breaks, and How to Detect Each One
Context failures look like a bad model. They are not. The four distinct failure modes, the symptom each produces, and the signal that tells them apart in production.
- about 2 months agoAI Coding
Three Guards Every Agent Loop Needs, and How Each One Fails
The loop is the easy part. What a step limit, a validator and a confirmation gate each miss — found by an independent review, not by the tests I wrote myself.
- about 2 months agoAI Coding
A Local AI QA Agent: Ollama, LibreChat, and Playwright MCP
Build a browser-automation agent that runs entirely on your machine — no API bill, no data leaving the laptop. Setup, the parts that break, and where it stops being useful.
- 2 months agoProgramming
Playwright 1.62.1 in Practice: The Changes That Matter for Automation QA
Practical Playwright 1.62.1 review: isolated retries, AbortSignal, WebP snapshots, API timing, passkeys, component testing, and MCP tooling for Automation QA.
