Financially Cooked.

Stop guessing where your money went. Financially Cooked is the reality-check expense tracker that roasts your bad spending habits.

Comment

Inspiration

It started with a simple, tragic glance at our bank accounts after a weekend of "treating ourselves." The numbers were grim. We realized that traditional budgeting apps are way too nice. "Oh, you overspent by $200 on food delivery? That's okay, try again next month!"

No. That doesn't work. Positive reinforcement is a myth. We needed an app that looks at our financial decisions and actively judges us. We needed an app with the emotional warmth of a deeply disappointed parent. Thus, Financially Cooked was born out of pure financial anxiety.

What it does

It is an expense tracker that thrives on your guilt and public humiliation. You log your purchases, and if you are spending money on things you absolutely do not need, the app ruthlessly roasts your life choices. It doesn't just track your budget; it attacks your ego so you stop impulse-buying junk.

How we built it

We built the guilt-inducing user interface using TypeScript and Next.js, ensuring every financial roasting is delivered seamlessly and at lightning speed. For the brain of the operation, we used Python to write the ruthless backend logic that categorizes your spending and calculates exactly how "cooked" you are. Finally, we hooked the whole thing up to a Supabase database. This ensures your terrible financial decisions are permanently etched into the cloud, meaning you can never hide from your past mistakes.

Challenges we ran into

Building a full-stack application in a weekend is basically a series of boss fights. Our first major hurdle was database synchronization. Our Python backend was evolving so fast that we ran into severe PostgREST schema cache errors and strict Row Level Security blocks in Supabase. We also accidentally recreated Zeno's Paradox while building the Splitwise debt settlement algorithm. Every time a user tried to pay someone back, it shifted the mathematical average of the group's total spending, making it look like they only paid half their debt! We had to engineer a simultaneous offset algorithm to balance the ledger without inflating the group average. Finally, in the last hour of the hackathon, Vercel had an identity crisis trying to build our Python backend instead of Next.js, and our last-minute Gemini AI calendar scanner got hit with a deprecated model API error right as we deployed. We had to hotfix the API version to 2.5 Flash with literally minutes on the clock.

Accomplishments that we're proud of

We are incredibly proud of building a fully functional, seamless full-stack application under tight hackathon time constraints. Getting a modern Next.js frontend to handshake flawlessly with a custom Python backend and a Supabase database is a massive win for the team. Beyond the code, we're proud of the UI/UX - it is clean, responsive, and perfectly delivers the exact toxic, guilt-inducing energy we originally envisioned.

What we learned

On the technical side, we leveled up our TypeScript skills and learned a massive amount about database management and API routing with Supabase and Next.js. We also figured out how to handle the quirky edge cases of passing complex JSON payloads between Python and JavaScript environments.

On a personal level, we learned that building an app specifically designed to bully you into saving money is both highly motivating and deeply stressful.

What's next for Financially Cooked.

Oh, we have plans to make it so much worse.

  1. The "Girl Math" Logic Engine: We are finalizing our "Girl Math" mode. When enabled, the app will try to use twisted, delusional logic to justify your terrible purchases (e.g., "If I pay with cash, it's basically free"), only to brutally pull the rug out and show you how much debt you're actually in.

  2. Multiplayer Sabotage: We're taking financial ruin social. We are building out the "Sabotage" feature, allowing users in the same group to deploy attacks on each other's dashboards when they catch their friends overspending. Peer pressure is a powerful financial tool.

  3. Bank API Integration (Plaid): Right now, users have to manually input expenses. Next, we want to integrate Plaid so the app can automatically scrape your bank account. You won't be able to hide your 2:00 AM fast-food runs anymore.

  4. AI-Powered Roasting: We plan to hook up an LLM to generate dynamically cruel, hyper-personalized roasts based on the exact item you just bought, rather than just general categories.

  5. Fixing the Bugs: We still need to iron out a few UI bugs and backend routing quirks. Right now, the app is so mean that it occasionally crashes out of pure spite. We need to stabilize it so the emotional damage is uninterrupted.

Built With

Share this project:

Updates