apr 18–19, 2026 · atlas
unseen corpora or it doesn’t count
8 corpora · 6 unseen during authoring

- timestamp
- on apr 19, the research draft checked whether the result held on repositories atlas had never seen.
- benchmark setup
- i mapped symbols, calls, and tests, then compared the agent against plain text tools before trusting the graph.
- suspicious result
- the first result looked too good, so i gave the judge code tools and added unseen corpora before trusting it.
- the pattern
- i turn an idea into a tool, measure it, look for the embarrassing counterexample, and keep whatever survives.
tags
- #code intelligence
- #evals
- #sqlite graph
- #mcp