polyglot-tutor / evals /README.md
blizzarman's picture
feat: M0 walking skeleton (app, service protocols, CI/CD, Space deploy)
00d5ae5
|
Raw
History Blame Contribute Delete
177 Bytes
# evals/
Evaluation harness (MLflow tracking on local SQLite, LLM-as-judge for generated exercises).
First real content lands with M1; per-milestone reports go to docs/evals/.