Inspiration

Online shopping has become part of our everyday lives, but finding the right product at the right price is not always easy. People often spend hours comparing products across different platforms like Amazon and Flipkart. Even after choosing a product, unexpected price changes, additional charges, and unreliable sellers can make online shopping stressful.

We wondered: What if AI could make shopping easier while also protecting users' money?

That question inspired us to build TrustCart AI.

What we built

TrustCart AI is an AI-powered shopping assistant designed to make online shopping smarter, easier, and safer.

Users can search for products, explore recommendations, compare available prices, and find products that match their needs and budget. Our goal is to bring product discovery and price comparison into one convenient shopping experience.

But we wanted to go beyond ordinary product recommendations.

We introduced a feature called the Transaction Consent Passport (TCP). It records the exact transaction details approved by the user, including the product, seller, quantity, and total amount. If an important detail changes before payment, the previous approval becomes invalid and the transaction is blocked until the user reviews and approves it again.

The platform also includes an AI shopping assistant, a Budget Guardian, transaction risk detection, and PayPal Sandbox checkout.

How we built it

We developed TrustCart AI using React and Vite for the frontend and Node.js with Express for the backend. We used a database to store product information, consent records, transactions, and audit logs.

We implemented an AI shopping assistant that understands user requirements and recommends suitable products. A rule-based risk engine checks important transaction conditions, while SHA-256 fingerprinting helps detect changes to approved transaction details.

We also integrated PayPal Sandbox to demonstrate a secure payment workflow in which the user remains in control of the final payment approval.

Challenges we faced

One of our biggest challenges was making sure that an AI-assisted shopping experience would not compromise user control over payments.

We had to think carefully about what should happen if a product price changes after approval, if a seller is unverified, or if a transaction exceeds the user's budget.

Another challenge was connecting the frontend, backend, database, and payment workflow while keeping the application reliable.

Working through these challenges helped us understand the importance of backend validation, transaction security, and clear communication between different parts of an application.

What we learned

Building TrustCart AI helped us improve our understanding of full-stack development, AI-powered recommendations, API integration, database management, and payment security.

Most importantly, we learned that building a useful AI product is not only about automation. It is also about giving users transparency, control, and confidence.

Our Vision

We want to make online shopping more intelligent without taking control away from the people who use it.

TrustCart AI – Shop with AI. Pay with Confidence.

Built With

  • ai
  • api
  • api-frontend-backend
  • api-payment
  • assistant
  • css3-website
  • express.js-backend
  • fingerprinting
  • html5-web
  • javascript
  • javascript-application
  • nlp-shopping
  • node.js-backend
  • paypal
  • react-frontend
  • rest
  • sdk-checkout
  • sha-256-transaction
  • sqlite-database
  • structure
  • styling
  • vite-frontend
Share this project:

Updates

Submission history