Markdown source

Harness Engineering: Building the Production Cage for Powerful Domain Agents

Conference Context

Session Description

Every agent is a while loop. The model takes strings in and produces strings out. We've all written it, debugged it, shipped it. And yet every team building agents is still re-inventing the same session management, truncation logic, tool wiring, and memory plumbing from scratch. The hard part is the harness: session isolation, context management, memory persistence, sandboxed execution, observability. The machinery that makes a model dependable in production. Most of the failures we see in deployed agents (context rot, premature completion, tool bloat) trace back to harness problems, not model problems. This talk covers what a harness actually does, why "harness engineering" suddenly showed up in engineering posts from everyone, and what changes when you stop building harnesses by hand. In live demos, we'll build the same agent three ways: hand-rolled Python, framework-generated, and fully managed through a single API call. Each level shifts the failure modes from infrastructure plumbing to engineering judgment, where the real questions are what context to preserve, when to verify, and how to keep an agent from finishing half the job and calling it done. The harness handles the machinery. You still have to engineer the behavior.

Media Evidence

Ship it! Building Production Ready Agents — Mike Chambers, AWS (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).

Evidence Graph

This evidence graph is generated from currently linked source material: official schedule text, related video pages, cached transcripts, visible slide text, dense/reconstructed slide pages, and AI slide-classification audits.

Media Signals

Agent Reading Notes

Use these signals to refine the synopsis, topic links, people/company context, and method notes. If a source is a related external video rather than an exact official recording, keep it framed as supporting evidence.

Transcript Status

Related video transcript availability: English auto-captions. Treat this as supporting context, not a recording of this exact scheduled session unless later confirmed. Not fetched yet.

People

Supporting Slides

Slide Evidence

Livestream Segment

Attendance Visibility

No high-confidence attendance icon signal is shown for this talk. The sampled video evidence was either low confidence, source-proxy-only, or did not expose a clear audience view.

Synthesis

Synthesized Breakdown

Mhm. Mhm. Mhm. Ladies and gentlemen, welcome to the AI Engineer World's Fair.

Speaker And Company Context

Topics Covered

Derived Links And Source Material

Novel Concepts / Clever Methods

Evidence Boundary

This synthesis uses the official schedule plus cached video transcripts. Official AI Engineer World's Fair San Francisco 2026 livestreams and cut videos are primary event video sources for transcript/slide evidence; external, historical, or speaker-matched videos remain supporting context unless manually verified as exact official event recordings.