I created this project for the H01 AWS + Vercel Hackathon (#H0Hackathon).

What I Built

ShopSmart AI is a conversational shopping assistant that helps users
plan events and DIY projects by recommending the right products instantly.

Try it live: https://shopsmart-ai-build.vercel.app

The Problem

Planning an event or tackling a DIY project means spending hours
searching across multiple sites. ShopSmart AI eliminates that friction
— describe what you need in plain English and get a smart product catalog instantly.

Tech Stack

AWS DynamoDB — product catalog (40 items, 8 categories, composite key: category + productId)
Vercel — frontend deployment and AI Gateway
Next.js + TypeScript — frontend framework
Vercel AI SDK — streaming chat with 5 custom tools
GPT-4o-mini — AI recommendations

How It Works

User describes their event or DIY project in the chat
AI asks clarifying questions
User confirms with "ok"
AI fetches real products from DynamoDB
Catalog populates with categorized product recommendations

DynamoDB Architecture

The table shopsmart-db uses a composite key:

Partition key: category (e.g. Decorations, Electrical, Plumbing)
Sort key: productId

On first query, AI recommendations are saved to DynamoDB.
Repeat queries serve from the database — faster and cheaper at scale.

What's Next

Affiliate integration with Amazon and Target
User accounts with persistent shopping lists
Mobile app
B2B white-label API for retailers

Conclusion

Building with AWS DynamoDB and Vercel was a great experience —
the Vercel native AWS integration via OIDC tokens made credential
management seamless in production.