Inspiration

Online shopping often requires users to spend considerable time searching for products, comparing specifications, evaluating prices, and staying within a fixed budget. Finding the right combination of products that satisfies individual requirements can be challenging.

We wanted to simplify this experience by building an intelligent shopping assistant that brings product discovery, comparison, budget optimization, and checkout together in one platform.

This inspired us to create BuyWise AI – Discover Smarter. Compare Better. Spend Wisely.

What it does

BuyWise AI is an AI-powered autonomous shopping assistant that helps users make smarter and more informed purchasing decisions.

Users can describe their shopping requirements in natural language, including their budget, preferred specifications, and priorities.

Its key features include:

  • AI-Powered Product Discovery: Uses Google Gemini to understand natural-language shopping requests and extract structured requirements.
  • Intelligent Recommendations: Filters and ranks products based on user preferences, specifications, and budget.
  • Product Comparison: Allows users to compare alternatives and evaluate product features.
  • Budget Optimization: Identifies suitable product combinations while respecting the user's spending limit.
  • Shopping Cart: Enables users to review selected products and validate the total purchase amount.
  • PayPal Integration: Supports PayPal Sandbox order creation, buyer approval, server-side payment capture, and purchase confirmation.

How we built it

We developed BuyWise AI using a full-stack architecture.

Frontend: Next.js, React, TypeScript, and Tailwind CSS provide an interactive and responsive shopping experience.

Backend: Python and FastAPI handle product discovery, recommendations, comparisons, budget optimization, cart validation, and payment operations.

AI Integration: Google Gemini API interprets natural-language shopping requests and converts them into structured requirements.

Recommendation Engine: A Python-based deterministic filtering and ranking system identifies suitable products using the extracted requirements and available catalogue data.

Budget Optimizer: Evaluates product combinations against the user's spending limit to help identify suitable purchases.

Payment Integration: PayPal Orders API Sandbox handles order creation, buyer approval, and server-side payment capture.

Deployment: The frontend and backend are deployed separately using Vercel.

Challenges we ran into

One of our main challenges was translating flexible natural-language shopping requests into structured product requirements. We addressed this by combining Gemini-powered intent extraction with deterministic filtering and ranking, keeping recommendations grounded in the available product data.

Another challenge was developing a budget optimization workflow that considers multiple products together instead of evaluating each product independently.

Integrating the complete PayPal checkout lifecycle was also challenging. We implemented order creation, buyer approval, server-side payment capture, and payment confirmation using PayPal Sandbox.

We also worked through deployment and frontend-backend integration challenges to ensure the complete shopping workflow functioned correctly in the live application.

Accomplishments that we're proud of

  • Built a complete AI-powered shopping assistant rather than a standalone product recommendation chatbot.
  • Integrated natural-language understanding with a structured recommendation engine.
  • Implemented product comparison and multi-product budget optimization.
  • Successfully integrated PayPal Sandbox checkout, including order creation, buyer approval, and server-side capture.
  • Deployed a working full-stack application with separate frontend and backend services.
  • Successfully tested the complete shopping journey, from product discovery to checkout confirmation.

What we learned

We gained practical experience integrating generative AI into a full-stack application and learned how to combine AI-driven intent understanding with deterministic business logic for reliable recommendations.

We also strengthened our understanding of REST API development, frontend-backend communication, recommendation systems, budget optimization, payment integration, and cloud deployment.

Most importantly, we learned how different technologies can work together to transform a simple shopping experience into an intelligent, end-to-end workflow.

What's next for BuyWise AI

We envision expanding BuyWise AI into a more comprehensive shopping platform.

Our future plans include:

  • Integrating live retailer APIs for real-time product information, prices, and availability.
  • Expanding the product catalogue across more categories.
  • Introducing personalized recommendations based on user preferences and shopping history.
  • Adding price tracking and deal alerts.
  • Improving budget optimization with more advanced constraints and preferences.
  • Supporting persistent user accounts, saved carts, and purchase history.

Our long-term goal is to make BuyWise AI a reliable intelligent shopping companion that helps users make better purchasing decisions while getting the most value from their money.

BuyWise AI – Discover Smarter. Compare Better. Spend Wisely.

Built With

Share this project:

Updates

Submission history