🫡 Inspiration

A 90% in Winnipeg isn't the same as a 90% in Vancouver. Canada has no national curriculum, so every school plays by its own rules. We built Axiom to bring the missing standard: a difficulty adjustment factor that makes grades mean the same thing.

🔥 What it does

1) Explore or ask
Use the map to filter schools by rating/rank, or ask our multi-agent system questions in plain English.

2) Grounded data retrieval
Our agentic system judges and contextualizes from our uploaded school datasets and ranking files, using tool-based (BackBoard.io) search and comparison.

3) Analyze test difficulty
Upload a test PDF in the Analyzer. The system routes it to the best agentic subject analyzer, estimates difficulty, and returns an adjustment factor.

4) Make better decisions
Compare schools, benchmark performance, and apply standardized adjustments through our agentic algorithm to make grading and planning more consistent across contexts.

😁 How we built it

Frontend:

  • Next.js (React + TypeScript)
  • Tailwind CSS
  • Framer Motion
  • Mapbox/MapLibre GL
  • Zustand

Backend:

  • Next.js API Routes (TypeScript)
  • Backboard SDK (assistant + tool orchestration)
  • Google Gemini API via Backboard (test analysis)
  • Local JSON/CSV data pipeline scripts
  • Supabase integration (project support tooling)

Built With

Share this project:

Updates