872 million tokens processed — 97.63% of them cached inputs. 5,222 tool calls. 88 issues closed. Zero...
I Was About to Spend Money on Ads. So I Made an AI Agent Test My Product for 29 Hours.
Related
You Are Describing the Symptom, Not the Problem
You ask why there are nulls in the report and get good advice about handling nulls. The nulls were never the problem.
Define once, cite everywhere
LLM-written specs restate your product slightly differently every time. Paraphrase is drift. One canonical rule + citations with content digests fixes it in 30 seconds.
23 TypeScript Tools for Making Software Explicit in the AI Era
In my previous articles, I argued that AI is changing the role of constraints in software...