MagicRadar

CLICK HERE: https://salmon-tree-032528610.7.azurestaticapps.net/ A corporate innovation intelligence tool built for Disney's Corporate Innovation team. MagicRadar enables analysts to track, evaluate, and generate AI-powered due diligence briefs on emerging technology companies across the media and entertainment landscape.

What it does

• Browse all 71 Disney Accelerator portfolio companies (2014–2025) with filtering by cohort year and sector
• Generate AI-powered due diligence briefs for any company — structured by overview, strategic fit for TWDC, technology assessment, opportunity & risk, and recommendation
• Research any emerging company outside the portfolio by entering a name and sector
• Built around how a corporate innovation team actually thinks — deal flow, business unit fit, investment thesis

Tech stack

• Frontend: Next.js + Tailwind CSS → Azure Static Web Apps
• Backend: Python FastAPI → Azure App Service
• AI: Anthropic Claude (claude-sonnet-4-20250514) for due diligence generation
• Cloud: Microsoft Azure (App Service, Static Web Apps, Resource Groups)

Architecture

The frontend connects to a FastAPI backend deployed on Azure App Service. The /api/brief endpoint takes a company name, sector, and optional notes, and returns a structured investment memo generated by Claude. The /api/companies endpoint returns the tracked portfolio.

Why I built this

The Disney Corporate Innovation team is responsible for identifying, evaluating, and executing on emerging technology opportunities across TWDC's business units. MagicRadar is a prototype of the kind of internal tooling that team would use — combining a curated portfolio database with AI-assisted research to accelerate due diligence.

Local development

# Backend
cd backend
source venv/bin/activate
python3 -m uvicorn main:app --reload

# Frontend
cd frontend
npm run dev

Built by

Althea Aguel — ECE Junior, Princeton University