alexgreensh/eval-genius
VerifiedResearchDocumented
eval-genius: Teach your agent to work with evals: WHEN you actually need an eval or benchmark, HOW to build one that holds up. See Jev references and source code.
Repository description: Teach your agent to work with evals: WHEN you actually need an eval or benchmark, HOW to build one that holds up, and how to read what it tells you. Deterministic-first, tool-agnostic.
Clone the repository and follow its README for install and usage. JevHunt has not run this project.
git clone https://github.com/alexgreensh/eval-genius.gitStars measure the whole repository, including work unrelated to Jev.
Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
Some of what an eval checks is a plain typed decision: real defect or not, which failure class, positive or negative sentiment, does this match the brand voice. A full reasoning model is overkill for those, and grading them by hand at volume is the real waste. Eval Genius can route exactly that residue to **Jev**, TypeSafe's hosted decision model, after your deterministic checks have run on everything first. You give Jev a typed question (yes/no, pick-from-a-list, or an ordinal score) and it ret
Evidence commit: dd1270441c2fbb75f85253571e2c74e7a4ad183c
Discovered through: github-search.
How we review →Suggest a correction · Claim this listing · Submit a project