6 Matching Annotations
  1. Last 7 days
    1. Layer 5: Trace evals. Evaluate the internal execution path: model calls, tool calls, handoffs, guardrails, retries, intermediate reasoning. They are the agentic equivalent of replaying the game tape: the final score matters, but the coach wants to know how the team played.

      这部分 trace的eval 需要借助API 分析

  2. Jan 2026
  3. Oct 2023
  4. Oct 2020
    1. You need to get out of the habit of thinking using quotes is ugly. Not using them is ugly! Why? Because you've created a function that can only be used interactively - it's very difficult to program with it. – hadley

      Does it seem like Hadley still stands by this statement after tidy evaluation from this article <Do you need tidyeval>

  5. Sep 2020