The Next Run Should Be Better

Official Schedule Context

Official Description

Agents generate a constant stream of experience through traces: tool calls, failures, corrections,

routing decisions, and user feedback. The challenge is identifying which parts of that experience

are worth remembering, and making those lessons available to the agent when it runs again. This talk

presents memory as an agent learning loop: capture traces, extract signal, and turn the right

lessons into durable context. We'll explore practical models for agent memory and discuss how to

build systems where the next run can be better than the last.

Related YouTube Video

No related AI Engineer channel video found yet.

Transcript Status

No official session recording transcript was found by exact title match on the AI Engineer YouTube channel during this run.

People

Notes