T
Oct 17, 2026Testing Free AI Coding Tiers: Four Myths and One Probe
A recurring scene in team chats: one engineer posts a screenshot of a free tier answering a genuinely...
Oct 17, 20266 min read0 reactions0 comments
Back to articles 
Oct 17, 2026
Oct 17, 2026
Tag archive
#testing
It Had Been Working Fine
Four people apologized. None of them was lying. The Screenshot It started with a...
Oct 17, 20263 min read0 reactions0 comments
Your Agent Wrote the Code and Its Test in One Turn. Green CI Proves Less
Labs have documented models special-casing and editing tests to pass. Treat agent-written tests as claims, protect a suite, flag turns that edit both.
Oct 17, 20268 min read0 reactions0 comments
T
Oct 17, 2026The Retry That Fired After the Stream Ended
The Retry That Fired After the Stream Ended A stream can succeed and still look failed to...
Oct 17, 20265 min read0 reactions0 comments
I
Oct 17, 2026Introducing... Becho: performance testing made easier
It's a self-hosted peformance testing benchmark web app that uses K6 under the hood. You can do...
Oct 17, 20261 min read0 reactions0 comments
I
Oct 17, 20268
Oct 17, 202689 of 190 reviewer subagents graded their own model's work. A PreToolUse hook has blocked 7 since; one walked around it
We want the subagent that grades our Claude Code agent's work to run on a different model family...
Oct 17, 202611 min read0 reactions0 comments
T
Oct 17, 2026The Stop List for Free Inference and Free Hosts
Most teams can name the day they adopted a free inference endpoint or a free server. Almost none can...
Oct 17, 20266 min read0 reactions0 comments
C
Oct 17, 2026Case Study: A Record/Replay Cassette That Makes an AI Agent Test Deterministic
The first failure in this build had nothing to do with model quality: the agent's test suite went red...
Oct 17, 20266 min read0 reactions0 comments
A
Oct 17, 2026A Green Pipeline That Ran Three of Five Shards Is Not Green
Your merge queue trusts a color. It should trust a coverage claim. A sharded CI run where shard 3...
Oct 17, 20266 min read0 reactions0 comments
I
Oct 17, 2026I Let Claude Code Write My Playwright Tests for a Month. Here's What Actually Worked
As a QA director at a Japanese IT company, I spent the last months building E2E test suites with an...
Oct 17, 20264 min read0 reactions0 comments
4
Oct 17, 202648-Hour Field Notes: Cutting a CI Log Down to the Failing Predicate
02:40, and the log was 12,417 lines long A deploy job went red on a branch nobody had...
Oct 17, 20267 min read0 reactions0 comments