Overview
What this challenge is about.
You receive 6 months of practice logs from 8,000 learners (item, timestamp, response correctness, response latency). Fit a learner-personalized forgetting model (logistic per-item + per-learner half-life regression is a strong baseline; HLR or a small recurrent network are acceptable upgrades). Simulate review-schedule decisions against held-out data and measure: items learned per minute, retention at 1/7/30 days, and learner-effort needed. Compare to the production hand-tuned baseline. Write a 4-page rollout proposal with A/B test design.
The Brief
What you'll do, and what you'll demonstrate.
Build a personalized forgetting-model-based spaced-repetition engine that beats the hand-tuned baseline on retention-per-effort.
Earning criteria — what you'll demonstrate
- Apply forgetting-curve models (HLR, half-life regression) to real behavioral data
- Personalize learning algorithms with per-learner parameters
- Run offline simulation as a substitute for online A/B in early stages
- Design an A/B test for a retention-sensitive product
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.
Machine Learning Engineer
Personalized scheduling models with offline simulation + A/B test plans are the MLE-shaped half of edtech recommendation systems.
This challenge sharpens
- personalization
- spaced-repetition
- regression-modeling
Data Scientist
Behavioral-data modeling with retention-per-effort framing is the daily data-science work at consumer-edtech companies.
This challenge sharpens
- behavioral-data
- regression-modeling
- ab-testing-design