jevPOST
LangChain author explains how Jev handles yes/no calls in agents
Chase relays ch3nweiii's point that many agent steps are binary choices rather than generation, and Jev answers them as typed questions with calibrated probabilities so frontier models focus on hard work.
Harrison Chase
@hwchase17Archived 21 days agoOriginally posted 2026.10.10
Harrison Chase @hwchase17Originally posted 2026.10.10
good framing from @ch3nweiii - a lot of agent steps are yes/no calls, not generation Jev answers those as typed questions with calibrated probabilities, so the frontier model sticks to the hard work where those calls sit in a harness: https://www.langchain.com/blog/building-a-harness-with-jev https://x.com/ch3nweiii/status/2108664035109470637
This post quotes another post: “Jev as a front-end decision model for agents, with cost figures”
x.com/hwchase17/status/2108777335650476163Archived 2026.10.11 10:06