Building a Coding Agent from Scratch

Learning Harness Engineering by Building a Coding Agent From Scratch

A complete guide to harness engineering, starting from first principles.

Oct 10 • Paul Iusztin

1:41:13

Will Your Next Change Break Your Agent?

Replay your agent's real failures after every prompt, tool, or model change

Sep 29 • Paul Iusztin

Agent Evals 101: Stop Guessing, Start Measuring

How to benchmark and regression-test any agent harness

Sep 22 • Paul Iusztin

Stop Babysitting Your Coding Agents

Run the harness remotely on Modal, triggered by your CLI, a webhook, or a nightly cron.

Sep 10 • Paul Iusztin

From 1 Bloated Context Window to 6 Scoped Subagents

Scope them with an agent catalog, run them in parallel, and fold the reports into 1 answer.

Sep 1 • Paul Iusztin

Context Engineering for Coding Agents

The 4 harness components that keep your context window high-signal.

Aug 25 • Paul Iusztin

From a Raw Shell to a Sandboxed Coding Agent

The guide to isolating your harness and safely executing its commands, locally or remotely.

Aug 18 • Paul Iusztin

The Bare-Bones Coding Agent Loop

One agent loop, 9 tools, and a terminal you can steer.

Jul 28 • Paul Iusztin

Building a Coding Agent From Scratch

Designing the harness around the model, from the agent loop to a remote swarm.

Jul 22 • Paul Iusztin

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts