Building a Codebase Classifier with Jev
A developer shares building a codebase classifier with Jev, suggesting it may solve overengineered code from agents, and asks what to test next.
Jev Test Ranking: 96% of Failing Tests in Top 10%
On 113 Defects4J bugs, the Jev model ranks test suites so that 96% of failing tests appear in the top 10%, outperforming BM25 and GPT-5.4 nano to fail builds faster.
Question: given a patch, can you rank a test suite so the tests that fail show up first? On 113 Defects4J bugs, how often is a failing test in the first 10%? BM25 → 73% GPT-5.4 nano → 92% Jev → 96% fail your builds quicker instead of waiting :)
· 0 likesOpen on X
More like this
Jev Enables LLM-as-a-Judge at 20–200x Speed and 40–400x Lower Cost
Jev was used to pick which model is AGI, offering 20–200x speed and 40–400x lower cost, making LLM-as-a-judge extremely fast and nearly free.
Testing apps with TypeSafe Jev & OpenCode
Filip shares a small demo showing how fast testing apps can be with TypeSafe Jev and OpenCode.
