Harness Engineering: How to Build a Software Factory — Dru Knox, Tessl
Oct 6, 2026AI Engineer
Summary
This post introduces the concept of 'harness engineering' (or 'loop engineering') as the discipline for transitioning from using coding agents to running a 'software factory.' It defines a software factory by autonomy, automation, and quality, and breaks down the work into inner, outer, and meta loops. The talk highlights the challenges of harness engineering, such as rapid field changes and localized logging. It proposes a three-layer architecture: a control plane, an agent-ready stack, and improvement loops, with Tessl's skills registry, Tessl Agent, and Tessl Launch presented as practical implementations to achieve these goals.
More from the Software Factory Engineering trail
- From Your Laptop to the Pipeline: Scaling Custom Agents with GitHub Copilot
- Generation Is Cheap, Review Is Expensive: How to Stop Shipping AI Slop — Gabriel Martinez, G2i
- Let Anyone at Your Company Ship Internal Apps with AI — Garrett Galow, WorkOS
- The Software Factory: From Bug Report to Production Code — Davis Palmie, Factory
- Why AI Agents Should Have Their Own Sandbox — Philipp Schmid, Google DeepMind