PayPilot AI

PayPilot AI combines intelligent product recommendations, real-time budget validation, and secure PayPal Sandbox checkout to make shopping smarter, more transparent, and budget-friendly.

Comment

Inspiration

We've all been in that situation where you have a strict budget and a list of things you urgently need to buy. Maybe you're a college student moving into an unfurnished dorm with only $150 left to get a desk lamp, a portable steamer, and noise-cancelling earbuds—or maybe your work laptop suddenly dies and you need a reliable replacement without wrecking your monthly finances.

Shopping online when money is tight is exhausting. You end up with twenty tabs open, trying to mentally balance product quality, warranties, delivery times, and prices. While AI chatbots are great at discussing what to buy, they can make mistakes with strict calculations—and no one wants an AI hallucinating prices when real money is involved.

That's why we built PayPilot AI: a budget-aware shopping assistant where AI handles the conversation and trade-offs, deterministic code checks the math, and PayPal handles the payment.

What it does

PayPilot AI helps people make smart, informed shopping decisions without losing sight of their budget. It guides users through a simple four-step journey: Describe, Compare, Check, and Pay.

  • Multi-Item Budget Optimizer & AI Chatbot: Users can describe their requirements in plain English, such as: "I have a strict $150 budget and need a desk lamp, portable steamer, and study earbuds." PayPilot AI helps interpret their needs, explore matching products, and explain the trade-offs between options.

  • Budget Guard: Our deterministic Budget Guard independently calculates bundle costs using catalogue prices and checks them against upcoming bills. It enforces the mathematical constraint:

$$\sum \text{Bundle Items} + \text{Upcoming 7-Day Bills} \leq \text{User Budget}$$

If a bundle exceeds the available budget, Budget Guard flags it and helps users explore cheaper alternatives.

  • Emergency Expense Navigator: Helps users compare replacement options for unexpected expenses based on available product information, warranty coverage, delivery speed, and total cost.

  • Price Context & Accessibility: Users can explore available price-history information, record upcoming bills, and use accessibility features such as high-contrast mode, scalable text, keyboard navigation, and voice controls.

  • PayPal Sandbox Checkout: Users can proceed through a PayPal Sandbox checkout workflow. Orders are updated based on backend verification of the payment process rather than simply assuming that opening checkout means a successful payment.

How we built it

As a three-person team, we divided responsibilities across the frontend, backend, AI functionality, payment integration, and testing. We coordinated our work through a shared GitHub repository and connected our modules through backend APIs.

  • Backend & Database: Developed the Flask and Python backend, including product catalogue APIs, bill management, order processing, and deterministic Budget Guard calculations. Integrated Neon PostgreSQL for persistent cloud database storage.

  • Frontend & User Experience: Built the responsive website using HTML, CSS, and JavaScript. Developed the product catalogue, budget dashboard, bill management interface, checkout experience, and accessibility controls.

  • AI, Payments & Integration: Worked on the intelligent shopping experience, product recommendations, PayPal Sandbox integration, and the complete order workflow, including buyer approval and payment verification.

All three team members collaborated on integration, debugging, testing, and preparing the final deployed application.

Challenges we ran into

  • Keeping AI Honest with Math: AI-generated responses can sometimes produce inaccurate calculations. We addressed this by separating AI reasoning from financial calculations. The backend Budget Guard independently calculates totals and checks spending limits.

  • Real Payment Verification: We wanted more than a simple payment button demonstration. We implemented a PayPal Sandbox workflow involving order creation, buyer approval, and backend payment verification.

  • Database Persistence: We needed our application data to survive backend restarts and redeployments. We integrated Neon PostgreSQL so that products, bills, and order records are stored independently of Render's local filesystem.

  • Working as a Small Team: With only three members, dividing responsibilities and integrating different modules was important. We focused on the essential functionality first and tested the complete application before deployment.

Accomplishments that we're proud of

  • Multi-Item Budgeting: Building a shopping experience where users can consider multiple product requirements together rather than evaluating every purchase separately.

  • Deterministic Budget Protection: Implementing a backend calculation system that independently checks product totals against the user's budget and upcoming bills.

  • PayPal Integration: Successfully demonstrating the PayPal Sandbox payment workflow, including buyer approval and payment capture.

  • Persistent Cloud Database: Integrating Neon PostgreSQL to retain application records across backend redeployments.

  • Complete Deployment: Bringing the frontend, Flask backend, database, and payment workflow together into one accessible web application.

What we learned

  • Deterministic Guardrails Matter: Generative AI is useful for understanding user requirements and explaining options, but financial calculations need predictable backend logic.

  • How Payment Lifecycles Work: Integrating PayPal Sandbox taught us the difference between order creation, buyer approval, and payment capture.

  • Cloud Database Integration: We learned how to connect a Flask backend to PostgreSQL and maintain persistent application data.

  • Teamwork and Scope Management: Working in a three-person team taught us to prioritize essential features, divide responsibilities, communicate clearly, and integrate our work efficiently.

What's next for PayPilot AI

  • Autonomous Group Travel & Event Co-Pilot: Expand the budget optimizer to support group budgets, shared expenses, and fair cost splitting.

  • Live Merchant & Price-History Feeds: Integrate live product feeds and verified price-tracking APIs.

  • Biometric & Passkey Approvals: Explore WebAuthn and passkey-based transaction confirmation for more convenient checkout security.

  • Smarter Financial Planning: Expand bill reminders and financial notifications to help users organize their expenses and plan their cash flow.

PayPilot AI – Think Smarter. Spend Better. Pay Smarter.

Built With

Share this project:

Updates

Submission history