Harness Engineering
A series on the emerging discipline of building infrastructure for AI agents — architectural oversight, constraint systems, and the shift from writing code to directing it.
Code generation is now infinite and essentially free. The bottleneck has moved from writing code to directing it — and that changes everything about how we build software.
Harness Engineering is the practice of building the infrastructure, constraints, and oversight systems that allow AI agents to operate safely and productively at scale. It's the difference between a powerful horse running wild and one guided by an experienced rider.
Articles in this Series
Part 1 — The Scarcity Inversion: Why Your Keyboard is Becoming Obsolete
Code generation is now infinite and free. The bottleneck has moved from writing code to architectural oversight — and that changes everything.
Part 2 — 1,500 PRs in 5 Months: The OpenAI Harness Benchmark
OpenAI's Harness team shipped ~1 million lines of production code with zero manual code — 1,500 PRs, 3.5 PRs/day/engineer, 90% time savings. Here's what it means for the future of software delivery.