How-ToAI AgentsNovember 6, 2026

Warp's Ben Holmes: agents hide mistakes, self-review fixes it

Read original source →youtube.com

Warp's Ben Holmes walks through logging every agent conversation with its tool calls and outputs in one place, then scoring them with LLM-as-judge rubrics for redundant tests, verbosity and efficiency.

People · Ben Holmes

1 source

More stories today

Open the live feed