← All popular talks

Popular talk #19

Harness Engineering: How to Build Software When Humans Steer, Agents Execute — Ryan Lopopolo, OpenAI

Ryan Lopopolo46:21

Synced transcript

Follow the talk

Intro
Harness Engineering
Scarce Resources
Guarding Quality
Prompting Agents
Q&A Intro
Harness & Tools
Adopting Agents
Scaling & Review
Tokens & Future
Outro

Community discussion

What did you agree with—or push back on?

Specific reactions make these talks more useful. Draft here, add the moment you’re discussing, then choose the direct-post pilot or the YouTube handoff.

Automated overview

What this talk covers

Ryan Lopopolo, a member of technical staff at OpenAI, argues that software engineers must shift to 'harness engineering' where agents produce all code and humans focus on delegation, system design, and guardrails. He describes how his team banned editors, using Codex with skills to manipulate apps, and reduced review friction by automating lint rules and review agents. Lopopolo explains that code is free but human time and model context are scarce, so teams should structure repositories for consistency and use 'garbage collection days' to eliminate recurring slop. He shares practical techniques like writing tests about source code (e.g., file length limits) and using review agents that inject prompts. His future vision involves agents autonomously executing long-horizon work based on token budgets and success metrics, freeing humans for higher-level activities.

This overview is derived from the transcript and has not been independently fact-checked by AI Engineer.

Chapters

  1. 0:00Intro
  2. 0:40Harness Engineering
  3. 5:11Scarce Resources
  4. 9:39Guarding Quality
  5. 13:23Prompting Agents
  6. 18:35Q&A Intro
  7. 19:52Harness & Tools
  8. 28:15Adopting Agents
  9. 33:39Scaling & Review
  10. 39:29Tokens & Future
  11. 45:58Outro