A pipeline to generate questions about estimating unknown quantities with AI.
You can play the game here.
With the current pipeline (with GPT-4o-mini and GPT-4o), generating 3,000 questions (filtered out of 50,000+ initial questions) costs approximately $20.
Create a file called .env containing OPENAI_API_KEY=<your OpenAI key>.
To run the pipeline: python src/main.py.
Move the generated file to to interface/backend/data.json directory and execute server.js with node. Finally, start a webserver in interface/frontend.