PayPilot

PayPilot helps users research products, compare prices, manage budgets, and make informed purchases with AI insights, spending rules, currency conversion, and user-approved PayPal Sandbox payments.

Comment

PayPilot – Smarter Shopping, Better Spending Decisions

Inspiration

Online shopping offers countless choices, but making the right purchase is not always easy. Users often have to visit multiple websites, compare products manually, calculate costs, and decide whether a purchase fits their budget.

We wanted to simplify this process by bringing product research, comparison, budgeting, and payment approval into one platform. This inspired us to build PayPilot, a smart shopping assistant that helps users make informed purchasing decisions while staying in control of their spending.

What it does

PayPilot guides users through a structured purchase workflow:

  • Describe: Users explain what they want to purchase using natural language.
  • Research: The assistant helps explore relevant products and available options.
  • Compare: Users can review product details, pricing, and alternatives.
  • Manage Budget: Spending rules help users evaluate purchases against their budget.
  • Review and Approve: Users review purchase details before authorizing payment.
  • Checkout: PayPal Sandbox demonstrates payment approval and capture.
  • Verify: The backend verifies the payment result before recording a successful transaction.

PayPilot is a purchase decision-support platform, not an online retailer. Its PayPal integration demonstrates a controlled payment workflow rather than placing orders with external stores.

How we built it

We built PayPilot using a full-stack architecture with the following technologies:

  • Frontend: React, TypeScript, Vite, and CSS.
  • Backend: Node.js and Express.js for API development and business logic.
  • Database: MongoDB Atlas for storing application data and transaction records.
  • Research Assistance: AI-powered support for exploring and comparing purchase options.
  • Payment Integration: PayPal Sandbox for payment approval, capture, and verification.
  • Currency Conversion: Frankfurter API for exchange-rate reference data.
  • Deployment: Vercel for the frontend and Render for the backend.

We separated the frontend experience from backend transaction handling to keep spending checks and payment verification on the server. A key design principle was to ensure that users explicitly approve payments before any capture takes place.

Challenges we ran into

One of our biggest challenges was integrating the PayPal Sandbox payment lifecycle. Creating orders, handling buyer approval, capturing payments, and verifying transaction status required careful coordination between the frontend and backend.

We also faced an issue where an already-approved PayPal order could prevent a new checkout attempt. We addressed this by implementing a recovery flow that checks the existing order and attempts to resolve the original transaction instead of unnecessarily creating another order.

Other challenges included handling API failures, maintaining consistent transaction states, coordinating frontend and backend data, and configuring environment variables across deployment platforms.

Accomplishments that we're proud of

  • Built a complete full-stack shopping research and budgeting application.
  • Integrated AI-powered research assistance into the purchase-planning workflow.
  • Implemented budget-aware purchase checks and explicit user approval.
  • Integrated PayPal Sandbox for payment approval and capture.
  • Added backend payment verification and transaction recovery.
  • Connected currency conversion to support international price estimates.
  • Successfully deployed the frontend and backend.

We are especially proud of bringing research, budgeting, and payment approval into one guided experience while keeping the user in control of every payment decision.

What we learned

Building PayPilot gave us practical experience in full-stack development, API integration, database management, payment workflows, and deployment.

We learned how important backend validation, reliable transaction handling, idempotency, and error recovery are when working with payment systems. We also gained a better understanding of integrating AI-assisted features into a real application without removing human control from important decisions.

Most importantly, we learned that a useful product is not just about adding features; it is about making the entire user journey reliable, transparent, and easy to understand.

What's next for PayPilot

We plan to continue improving PayPilot by:

  • Enhancing product research and comparison capabilities.
  • Introducing more detailed spending insights and budget tracking.
  • Improving the accuracy and usefulness of purchase recommendations.
  • Expanding currency and pricing support.
  • Improving the overall user experience and transaction recovery process.
  • Exploring additional payment integrations and more realistic checkout workflows.

Our long-term vision is to make PayPilot a reliable shopping companion that helps people research better, spend wisely, and make confident purchasing decisions.

Built With

Share this project:

Updates

Submission history