Eve (Cal Hacks 2026)
event finance agent that handles hackathon reimbursements, prizes, and cross-border tax, with a human controller releasing every payout
- Timeline
- Fall 2026
- Stack
- Next.js, TypeScript, PostgreSQL, Drizzle, Claude, Vercel Workflow
- Links
Inspiration
Evan, one of my teammates, won a prize at TreeHacks. The organizers first tried to ship the physical prize, then found they couldn't for tax reasons, because he's a Canadian citizen. They switched to a wire. Getting the right tax form, the right withholding, and a working international transfer took about nine months.
Nobody did anything wrong. The organization simply had no system that knew, at the moment he won, that a Canadian winner needs a W-8BEN, a withholding decision, and a cross-border payout rail. The same failure repeats at every event: receipts in other currencies, caps nobody checks until after the flight is booked, tax forms collected late, and a university or fiscal sponsor that holds the bank account.
We asked: what if one permissioned agent did the travel coordinator's, accounts payable clerk's, and tax preparer's jobs, while a human still decides when money moves?
What It Does
Eve runs the money side of a hackathon. Organizers set budgets and rules, attendees submit receipts and claim prizes in a portal, and Eve prepares every payout with its evidence. Nothing is released without a human controller. (Money movement is simulated, and tax documents are drafts.)
- Overview for organizers. Budget left, payouts awaiting release, what is blocked and why, and the prize pipeline before the closing ceremony.
- Travel and expenses. Policy-aware flight suggestions, receipt reading, caps, currency conversion, and duplicate and fare-anomaly checks.
- Cash and physical prizes. For each item prize, Eve compares shipping, buying locally, and a cash equivalent. Team splits need member confirmation.
- Tax profiles and withholding. Onboarding collects residency and W-9 or W-8BEN details; withholding, fees, and FX are computed before approval.
- Controller release. Every proposal discloses gross, withholding, fees, and expected net. Only the controller can release it; the organizer and the agent cannot.
- Fraud guards. Claimed fares are compared with peers and market prices, travel claims from hackers with no project on record are escalated, and a bank change near payout freezes the payment until it's confirmed.
- Slack approvals, sponsor handoff, and audit. Controllers approve by DM, Eve fills a fiscal sponsor's portal with computer use and stops before Submit, and an immutable double-entry ledger backs every number.
How We Built It
The frontend is Next.js 16 (App Router), React 19, TypeScript, Tailwind CSS 4, shadcn/ui, Base UI, and TanStack Table. The backend uses server actions and route handlers over Postgres (PlanetScale) with Drizzle ORM, Better Auth, Zod, Vercel Workflow for durable jobs and cron, and pdf-lib for documents.
On the AI side, Claude (via the Vercel AI SDK and Anthropic SDK) reads receipts with structured extraction and powers the event review agent, and Claude computer use on an E2B desktop enters sponsor packets. Policy, tax, and money calculations are deterministic engines: the model never decides an amount.
Challenges
- Autonomy without a superuser. The agent is a role with permissions enforced in code, not in the prompt. It can propose and hold, never release.
- Sources that disagree. Receipts, card charges, policy, the sponsor's ledger, and the payee's bank each say something slightly different, and Eve has to say which to trust and why.
- Cross-border tax. Cumulative per-payee thresholds, default 30% withholding without a valid treaty claim, and Canadian rules we kept explicitly unverified instead of guessing.
Accomplishments
- A seeded 300-person event where the measured scoreboard caught 40 of 40 planted problems, with no false positives and no wrongful payouts
- A balanced, immutable ledger where every number traces back to its decision record
- Computer use that fills a realistic sponsor portal end to end and leaves the final Submit to a person
What We Learned
The hard part of a finance agent isn't the model; it's deciding what the model is allowed to do. Keeping calculations deterministic, rules versioned, and release with a human let us automate the routine work without trusting generation with money.
My Contributions
I built Eve's airfare fraud detection: peer fare anomaly checks, market price snapshots, and nightly cheapest-fare tracking, so an inflated flight claim stands out against what others paid and what the route actually cost. I also built attendee verification against project submissions, which escalates travel claims from hackers with no project on record. On the interface side, I designed the Daylight theme and the Overview home. My teammates covered product and architecture, the ledger, tax, payout, and policy engines, receipt reading, and the agent (Evan), and the Slack app, computer use on the sponsor portal, and the Lamplight dark theme (Bernice).
What's Next
- Live payment rails beyond test mode, and verified flight booking
- Verified Canadian rules and treaty handling
- Official IRS and CRA forms and e-filing
- Email verification and invitation delivery for real organizations