Lessons and explainers that make AI make sense.
Google engineers measured how stale bug-reference comments pull AI coding agents off task, and removed thousands of them in six months.
audiotale.net team6 min read
audiotale.net team6 min read
audiotale.net team8 min read
audiotale.net team7 min read
audiotale.net team6 min read
A plain explanation of the LLM KV cache, prefill versus decode, paging, and how a sizing mismatch let one user's data reach another in a recent incident.
audiotale.net team6 min read
A plain guide to formal verification: what a machine-checked proof contains, how it differs from tests and fuzzing, and the limits to check whenever software is called formally verified.
audiotale.net team7 min read
Anthropic says Claude wrote machine-checked proofs for 123 open-source libraries in a few weeks and found hundreds of bugs, a claim that formal verification can now scale; the authors list the limits.
audiotale.net team7 min read
A step-by-step guide to prefix caching for LLM requests: what it costs, where to put breakpoints, what breaks it, and how to measure your hit rate.
audiotale.net team7 min read