Solo-hackathon • Full-Stack • October 2025
Auddit — Government Spending, Mapped
Every year, the government distributes trillions of dollars to hospitals, contractors, nonprofits, childcare centers, and countless organizations in between. By law, nearly all of it is publicly disclosed. But publicly available and actually findable are two very different things. The data lives scattered across government portals and was built for policy analysts not the person who wants to know what's happening in their own town. I built Auddit into a full-stack app that pulls it all into one map, lets citizens post publicly what they know, and submit evidence privately that can be passed as legal cases.
The interactive map is the center of the product. Every dot is a real organization receiving real taxpayer dollars. The data is continuously ingested from federal open APIs and supplemented with state-level spending portals where most of the subsidy details live.
Every recipient gets a full profile with their total funding received, number of awards, and a complete itemized history of every grant.
Recipients are organized into dynamic lists generated by state, funding agency, and grant type. Users can also build their own custom lists and filter by geography, program, funding amount, or activity to track what matters to them over time.
Conversations: Each recipient has a public comment thread where locals, journalists, and concerned citizens can post observations and add context. Posts are interactive like reddit and made public but identity can be private.
Evidence: Users can submit formal evidence tied to a specific recipient - attaching documents, photos, or records with any context. These submissions are always kept private and can be passed along to in-house legal or an external firm. Our AI scans recipient activity, funding patterns, and submitted evidence to automatically surface anomalies and package findings into a lawyer-ready case for Qui Tam filing.
Qui Tam: A 160-year-old law that lets private citizens file fraud suits on behalf of the government and keep a share of whatever is recovered. Auddit is the investigation layer that precedes it. In 2024 alone, Qui Tam suits recovered $2.9B
The initial build is done and live. To scale this, it requires distribution, content seeding, and moderation infrastructure. The political moment for this has never been stronger, and the legal mechanism to turn citizen findings into real consequences is already proven. It's waiting for the right team to execute.
