jevVIDEO
Kalmia uses Jev to pick key steps from long agent run logs
Kalmia uses Typesafe Jev to select the steps worth analyzing from thousands of steps in agent run logs before passing them to its analysis harness, finding the same bugs on real customer runs at 9x lower cost and 6x faster.
Calvin Huang
@calvinhhuanggArchived 24 days agoOriginally posted 2026.10.08
Calvin Huang @calvinhhuanggOriginally posted 2026.10.08
Long-running agents can log thousands of steps per run. Analyzing every step to find what went wrong is slow and expensive. @kalmiadev has been experimenting with @typesafeai Jev to help choose which steps matter for our harness to analyze. Same bugs found on real customer runs, 9x cheaper and 6x faster. Try now http://kalmia.dev
x.com/calvinhhuangg/status/2107871071454326933Archived 2026.10.08 10:09