World Congress 2026 North America
Agents Can't Iterate Against Tests That Lie
Rocky Warren
Senior Staff Software Engineer at Clipboard
World Congress 2026 North America
- Discuss this with your agent
- Open in Claude
- Open in ChatGPT
There's no dark factory without better software verifiers
September 23–25, 2026
World Congress 2026 North America
September 23–25, 2026 · San José, CA
Attend in person
Get ticketsWatch remotely
Watch live with Pro
Pro
Can’t make it to San José? Watch this session live with Pro. You also get:
- All full videos, bookmarks, and playlists
- World Congress livestreams
What this session covers
While the “do you have to read the code” debate rages on, we’ve been hard at work figuring out how to get models to write code that gets better over time, not worse. We’ll discuss why current model training approaches and benchmarks checking “do the tests pass” are not enough, and how we’re approaching evaluating the long-term-quality of LLM-generated code.
Related talks at this congress
World Congress 2026 North America
Your Evals Passed. Your Agent Just Emptied a Database.
Tejas Pravinbhai Patel
IEEE Award-Winning Researcher | Best Keynote Speaker | Sr. Software Engineer at Amazon | AI Systems & Agent Architect
World Congress 2026 North America
When Humans Stop Writing Code: Rethinking Languages, Compilers, and Responsibility
Simon Auer
Organizer of flutter vienna meetup and CEO of marqably
World Congress 2026 North America
Evals Are Infra: Building AI Systems Developers Can Actually Trust
Phoebe Wang
Member of Technical Staff at OpenAI
