Harness Engineering: Building the Production Cage for Powerful Domain Agents
Conference Context
- Date/time: 2026-07-01 · 12:05pm-12:25pm
- Track/room: Harness Engineering · Main Stage
- Speaker(s): Mike Chambers
- Session type/status: session · confirmed
- Track: Harness Engineering
- Room: Main Stage
- Session type: session
- Status: confirmed
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).
- youtube I2cbIws9j10 transcript — full cached transcript markdown for the related YouTube source.
- Source video:
youtube-HT4l0DeP69I - Slide deck: Dense Slides: Ship it! Building Production Ready Agents — Mike Chambers, AWS — 2 visible slide image(s); 2 HTML recreation(s).
- Additional slide evidence: Slides: Ship it! Building Production Ready Agents — Mike Chambers, AWS, Reconstructed Slides: Ship it! Building Production Ready Agents — Mike Chambers, AWS
- Slide-derived themes for
youtube-HT4l0DeP69I: mike, chambers, advocate, engineering, generative, real, world, applications. - Source video:
youtube-I2cbIws9j10 - Slide deck: Dense Slides: WF26: Harness Engineering & Startup Battlefield ft. Garry Tan, Mike Krieger, @t3dotgg , DSPy — 11 visible slide image(s); 11 HTML recreation(s).
- Additional slide evidence: Slides: WF26: Harness Engineering & Startup Battlefield ft. Garry Tan, Mike Krieger, @t3dotgg , DSPy
- Slide-derived themes for
youtube-I2cbIws9j10: context, window, selects, response, facts, retry, coerce, rollback.


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
youtube-HT4l0DeP69I— 10 slide-derived text signals- Slide-derived themes for
youtube-HT4l0DeP69I: mike, chambers, advocate, engineering, generative, real, world, applications. - Evidence links for
youtube-HT4l0DeP69I: youtube HT4l0DeP69I, youtube HT4l0DeP69I slides, youtube HT4l0DeP69I dense slides, youtube HT4l0DeP69I reconstructed slides youtube-I2cbIws9j10— 91,792 transcript words; 7 slide-derived text signals- Transcript signals for
youtube-I2cbIws9j10: code, model, back, system, well, first, today, even. - Slide-derived themes for
youtube-I2cbIws9j10: context, window, selects, response, facts, retry, coerce, rollback. - Evidence links for
youtube-I2cbIws9j10: youtube I2cbIws9j10, youtube I2cbIws9j10 transcript, youtube I2cbIws9j10 slides, youtube I2cbIws9j10 dense slides
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
- youtube HT4l0DeP69I slides — extracted from the related public AI Engineer video.
Slide Evidence
- Slide-only cropped deck: youtube HT4l0DeP69I dense slides (2 viable slide images).
- Related slide/OCR pages:
- youtube HT4l0DeP69I dense slides
- youtube HT4l0DeP69I reconstructed slides
- youtube HT4l0DeP69I slides
- Slide-derived terms:
models,amazon,microsoft,bedrock,model,prompt,mike,system,master,roll,claude,chat,select,nova,learn,world,ate-wf-2025-demos,grand
Livestream Segment
- Watch in livestream at 03:14:28 — WF26: Harness Engineering & Startup Battlefield (Day 3).
- Match basis: speaker and title; timed captions matched Mike Chambers, engineering, harness.
- Confidence: high automated match; prefer a dedicated cut-video recording when one exists.
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
- Mike Chambers — Senior Developer Advocate for Generative AI at Amazon Web Services (AWS).
Topics Covered
Derived Links And Source Material
- youtube I2cbIws9j10 transcript — transcript markdown; source cache
raw/sources/youtube-livestream-transcripts/I2cbIws9j10.txt(91,792 words). - youtube HT4l0DeP69I — related YouTube source page.
- youtube HT4l0DeP69I slides — slide evidence.
- youtube HT4l0DeP69I reconstructed slides — slide evidence.
- youtube HT4l0DeP69I dense slides — slide evidence.
- youtube I2cbIws9j10 — related YouTube source page.
- youtube I2cbIws9j10 slides — slide evidence.
- youtube I2cbIws9j10 dense slides — slide evidence.
Novel Concepts / Clever Methods
- Agent-Ready Accessibility — Designing for agents and designing for accessibility converge around explicit structure, reachable controls, and understandable state.
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.