Decision Arena

• 

  Dashboard

• 

  architecture diagram

About the Project

Inspiration

Business decisions like changing prices or launching new features are often made with limited data and involve significant risk. I wanted to build a tool that lets businesses test these decisions in a virtual market before making them in the real world.

What it does

Decision Arena uses AI to simulate how virtual customer segments—representing different demographics, behaviours, and preferences—react to a business decision. It predicts metrics like revenue impact, churn, customer sentiment, risks, and provides a recommendation, while also comparing similar past simulations.

How I built it

I built the application with the help of AI, especially for the UI using Vercel.

Challenges

The biggest challenges were integrating Amazon Bedrock Mantle, designing prompts that consistently returned structured outputs, and making the simulations feel realistic and useful.

What I learned

I learned how to build an end-to-end AI application using AWS services, integrate generative AI into a production-style web app, and design a scalable architecture for AI-powered business simulations.

AWS Database Used: Amazon Aurora PostgreSQL

Built With

• amazon-aurora-postgresql
• amazon-bedrock-mantle
• better-auth
• drizzle-orm
• google-gemma-3-12b-it
• next.js-15
• react
• tailwind-css
• typescript