AI Shopping agent

Inspiration

We’ve all been there opening ten tabs just to buy one item. Searching, comparing, checking reviews, and finally going through a long checkout. We wanted to fix that. The idea was simple: what if shopping could feel as easy as texting a friend who does it all for you? That’s how the AI Shopping Agent was born.

What it does

Our AI Shopping Agent is like your personal smart shopper. It chats with you naturally, answers product questions, adds things to your cart, finds deals, and even places orders for you — all in one smooth conversation.

How we built it

We used Python (Django REST) as the backbone and combined it with Amazon Bedrock AgentCore to bring our AI to life. I set up the Bedrock AgentCore Gateway, my teammate handled the Lambda functions, and another teammate configured the AI Bedrock Agent to make the conversations truly intelligent.

Challenges we ran into

Integrating everything across multiple AWS services wasn’t easy. We struggled with syncing user sessions, maintaining context, and ensuring the responses felt natural and fast. But every late-night bug fix was worth it.

Accomplishments that we're proud of

Seeing our AI handle an entire shopping session from “What’s the best phone?” to “Order it for me” was magical. We’re proud of how well we collaborated, learned new tools, and actually built something that feels futuristic.

What we learned

We learned how to connect Bedrock, Lambda, and Django into one ecosystem. More importantly, we learned how small, clear teamwork can turn an idea into a working product.

What's next for AI Shopping Agent

We want to make it even more personal voice support, smart recommendations, and integration with real stores. Our goal? To make online shopping feel like having your own personal shopper, powered by AI.

Built With

• amazon-bedrock-agentcore
• amazon-web-services
• amazonbedrockagentcore
• aws-lambda
• awsbedrock
• bedrock
• django
• lambda
• python