Hacklytics 2026 - Finance / Data Science Track
CartPilot is an autonomous "GPS for your wallet." Instead of linking to a bank (privacy-first), users input their financial goals (e.g., "Save $5k for Japan"). When a user scans a shopping cart, the AI calculates the opportunity cost of the purchase against their goal, runs a stochastic liquidity simulation to prevent overdrafts, and identifies neighborhood price arbitrage for essential items.
- Goal-Centric Onboarding: Set specific goals and current liquid cash
- OCR Receipt/Cart Parser: Extract items, prices, and categories from images
- Goal Impact Engine: Mathematical model projecting purchase impact on goal completion
- Monte Carlo Liquidity Forecast: 1,000-run simulation with safety confidence score
- Multi-Objective Split-Pay Optimizer: Linear programming for optimal payment strategies
- Neighborhood Arbitrage: Fuzzy matching for cheaper local alternatives
- Decision XAI: SHAP-based explanations for payment strategy choices
- Next.js (App Router)
- Tailwind CSS
- Lucide React (Icons)
- Recharts (Data Visualization)
- Python (FastAPI)
- AWS Lambda (Serverless logic)
- RapidFuzz: Product name normalization and arbitrage logic
- SciPy / Pyomo: Multi-objective optimization
- NumPy / Pandas: Monte Carlo simulations
- InterpretML / SHAP: Explainable AI
- Amazon Textract: OCR extraction
- DynamoDB (Card Rewards)
- Supabase/PostgreSQL (Price data)
- Vercel (Frontend + Python Serverless Functions)
-
Install dependencies:
npm install
-
Run the development server:
npm run dev
-
Open http://localhost:3000 in your browser.
src/app/- Next.js app router pagessrc/components/- Reusable React componentspublic/- Static assets
/- Home dashboard/onboarding- Goal setting/scan- Cart/receipt upload/analysis- Goal impact analysis/forecast- Liquidity simulation
- Implement backend API endpoints
- Add OCR processing for receipts
- Integrate Monte Carlo simulations
- Build payment optimization logic
- Add SHAP explanations
- Deploy to Vercel