Build a Math Intelligent-Tutoring Assistant for High Schoolers
Overview
What this challenge is about.
You receive: a curated set of 40 algebra problems with worked solutions, the company's pedagogy rubric ('hint, don't reveal' principle), and a baseline 'just answer' chatbot for comparison. Build a tutoring assistant using a large language model with a structured hint policy (e.g., 3-tier hints: nudge, partial step, full step) and a refusal pattern when students push for the answer. Run a 5-student blinded eval comparing learning gains and frustration on a 5-problem post-test. Deliver the prototype, eval results, and a pedagogy memo.
The Brief
What you'll do, and what you'll demonstrate.
Build and evaluate a Socratic math-tutoring assistant that hints rather than reveals, and beats a vanilla chatbot on learning gains.
Earning criteria — what you'll demonstrate
- Apply LLMs to intelligent-tutoring use cases with a structured hint policy
- Design prompt patterns that encode pedagogical principles
- Run a small mixed-methods evaluation with student testers
- Communicate trade-offs between answer-quality and learning-quality
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Engineer
Shipping an LLM tutoring product with a measurable pedagogy policy is the AI engineer project edtech startups hire on the strength of.
This challenge sharpens
- llm-agents
- intelligent-tutoring
- evaluation-design
AI Product Designer
Designing the hint UX and balancing helpfulness vs. learning quality is the AI product designer's mandate in edtech.
This challenge sharpens
- user-research
- evaluation-design
- pedagogy