Intel / video

How to Build a Software Factory That Improves Itself - Zach Lloyd, Warp

Why it matters

Frames agent infrastructure as code with LLM-judged run scoring and per-team benchmarks, so you can compare model cost and quality on your own tasks.

Key takeaways · AI-distilled

Zach Lloyd says companies are moving from engineers prompting agents on their laptops to cloud software factories, and argues the factory should be treated as infrastructure: open, compatible with existing tools and defined in code.
Warp Factories gives visibility into every run and scores runs with an judge, and a self-improvement loop proposes fixes to the factory itself.
Warp builds benchmarks from a team's own past tasks, so the team can compare cost and quality across models on the work it actually does.

Terms in this piece · Glossary

LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
sandbox — An isolated environment where AI-generated code or agent actions run without being able to touch anything real.