1st Place at AGI House AI4Healthcare Hackathon

We took 1st place at AGI House's inaugural AI4Healthcare hackathon with Proudfoot, an AI-powered search engine for public NGS data.

What we built

Proudfoot is a search engine that lets researchers find and rank public RNA-seq datasets using natural language queries. Instead of manually browsing GEO or SRA with keyword filters, you describe what you're looking for (a disease, a tissue type, a treatment condition) and Proudfoot retrieves the most relevant datasets.

It's the kind of tool that turns a multi-hour literature and database trawl into a single query.

The result

Out of all the teams competing, we took home 1st place. The judges saw what we saw: the barrier to using public genomic data isn't access, it's findability. There are millions of publicly available datasets, but discovering the right one for your research question is still surprisingly hard.

What came next

Proudfoot evolved into what is now Omabit. The core insight, that natural language is the right way to search and interact with genomic data, became the foundation of our product.

A huge thank you to AGI House and the sponsor Mithrl for putting together a fantastic event. The AI4Healthcare hackathon brought together an incredible group of builders at the intersection of AI and biology, and we're grateful to have been part of it.