# feeRnt0team

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [fee Rnt](https://cerebralvalley.ai/u/feeRnt)
- **GitHub:** https://github.com/feeRnt/whofundsit/
- **Demo video:** https://youtu.be/2zoHTMV4IFA
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4987

Many people want to know, when making a purchase, buying a subscription, or reading a medical study, what the primary incentives are behind the service or who are its prime beneficiaries. The best way to understand this information is to see what funding is behind any company, organization or even an educational institution.

This webapp is an interactive global corporate financing map that showcases the flow and source of money, with bigger graphical representation to places with more funds, e.g., BlackRock. Including hedge-funds to bring out the scale of debt, the sources for this mapping were stocks, equity, large donations and the like, and AI - Open-Weighted Muse Glimmer - is used to process hard-to-map business addresses on the map, and to test the soundness of the data collected using a random methodological test on our outputs.

 WhoFundsIt is an interactive global map of who funds whom - companies, banks, hedge funds, and universities appear as dots sized by money, with arcs showing investment flows, grants, and ownership between them. Click any dot for its value, address, and SEC evidence links.

  The problem it solves: when you buy a product, subscribe to a service, or read a medical study, the incentives behind it are buried in filings and databases no human browses for fun. WhoFundsIt combines public sources, e.g., SEC EDGAR (company facts, 13F
holdings), Wikidata, GLEIF, OpenFIGI, OpenAlex database, plus geocoded addresses, into one picture where the scale of money is instantly visible (why BlackRock dominates the frame). A stochastic verification audit scores the data against live sources and shows
a "% verified" badge, so you can see how much to trust what you're looking at (though unfortunately it could not make the final cut for the demo).

To me, the most amazing part of the project was the resources available to solve the problem. I had only had this idea about 2 or 3 weeks go on a random whim, and in the spirit of the hackathon (and Muse Code), it's a real project now. But I think there is so much more potential. Financial mapping of really small companies, banks, and private companies was not possible, but the presence of orgs like that at OpenAlex excited me, and showed me that my work is valuable. It's an amazing thing that we humans desire, and need, big data to understand the world around us, and I think my app definitely helps fill in a hole of both practicality and enjoyment.

The Stack: TypeScript + Express + Postgres/Prisma backend, deck.gl/MapLibre globe frontend, open-weight Glimmer workers on Modal for parsing addresses that are too unstructured for geocoding, and for audit commentary (live, daily feedback on the visualization and correctness of the data presented in the app).  While it's not possible (nor realistic) to run AI agents to parse every datum as it was acquired from primary sources, what we can audit is faults in the implementation, which I felt was an amazing idea for AI projects in general and mine.

## More from Meta Global AI Developer Hackathon

- [SnapQuest](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4984)
- [Schema](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4985)
- [david_xu_sole](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4986)
- [Dodona Agent Kernel](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4988)
- [Eclipse](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4989)
- [handoff.lol](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4990)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4987. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt