# Second_Loop_Engineering

- **Event:** [The Harness Engineering & Model Wrangling Hackathon](https://cerebralvalley.ai/e/mongodb-nyc-hackathon)
- **When:** Sat, Sep 26 at 9:00 AM – 10:00 PM (EDT)
- **Where:** New York, NY
- **Team:** [Chris Gibbs](https://cerebralvalley.ai/u/TallGibbs)
- **GitHub:** https://github.com/TallGibbs/race-engineer-adaptive-harness
- **Demo video:** https://drive.google.com/file/d/11EwE4t3Dj1k2Y88wV-x_tpCtBH8LacND/view?usp=drive_link
- **Gallery:** https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/23

In agent-run analytics and engineering workflows, outputs must cite their evidence; when they fail, teams tune prompts and checks by hand without a fixed measure. This project automates that loop for any such workflow. Three agents (data engineer, statistician, lead engineer) form an agentic engineering team inside a harness that learns from its recorded runs. Each improvement cycle finds root causes and proposes changes to its checks, stages, and what each agent sees. A fixed evaluator and five acceptance rules, set before any code, decide what stays. MongoDB Atlas is its memory: events, proposal history, and lessons retrieved by Atlas Vector Search.

## More from The Harness Engineering & Model Wrangling Hackathon

- [Ada](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/20)
- [PolicyPilot](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/21)
- [Moss](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/22)
- [devrandom.co](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/24)
- [Antibody](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/25)
- [Blanko](https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/26)

---

Markdown version of https://cerebralvalley.ai/e/mongodb-nyc-hackathon/hackathon/gallery/23. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt