# Team Tesfa

- **Event:** [The Harness Engineering & Model Wrangling #SFTechWeek Hackathon](https://cerebralvalley.ai/e/mongodb-sftw-hackathon)
- **When:** Sun, Oct 11 at 9:00 AM – 10:00 PM (PDT)
- **Where:** San Francisco, CA
- **Team:** [Abel Tesfa](https://cerebralvalley.ai/u/abelSF)
- **GitHub:** https://github.com/abeltesfa/longview
- **Demo video:** https://youtu.be/snZA3PoUrKU
- **Gallery:** https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/9

AI agents that write recurring business reports start from zero every week, so they can't say "third week in a row" or "the fix you tried didn't work." Longview gives an analyst agent persistent memory: each week it forms hypotheses about why revenue moved, tests each against live data in MongoDB Atlas, and reinforces, contradicts, or supersedes its beliefs. A self-improvement loop scores every run on hard metrics (no LLM in the scoring), reflects on its own failures and Raindrop telemetry, and writes a new playbook version (rules, guardrails, parameters), accepted only if it scores strictly higher.

## More from The Harness Engineering & Model Wrangling #SFTechWeek Hackathon

- [Procurator](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/6)
- [AgentShield](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/7)
- [SuperBid](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/8)
- [Routemind](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/10)
- [before-the-drop](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/11)
- [AIZOYA](https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/12)

---

Markdown version of https://cerebralvalley.ai/e/mongodb-sftw-hackathon/hackathon/gallery/9. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt