Overview
What this challenge is about.
You receive 5 years of hourly residential-segment demand, hourly weather data (temperature, wind, irradiance), and a calendar of public holidays. Build a probabilistic forecaster that emits a P10/P50/P90 forecast for each of the next 24 hours. Backtest on the last 12 months with a walk-forward split, focused on the 4 morning peak hours (06-10). Beat the incumbent gradient-boosted-tree baseline by 5 percent on mean pinball loss for those 4 hours. Deliver a trading-team-ready memo.
The Brief
What you'll do, and what you'll demonstrate.
Beat the incumbent peak-hours demand forecast by 5 percent on mean pinball loss with a probabilistic, weather-aware model.
Earning criteria — what you'll demonstrate
- Build a probabilistic time-series forecaster (not just point estimates)
- Apply walk-forward backtesting correctly on long horizons
- Engineer weather and calendar features that genuinely help
- Communicate model uncertainty to a trading-floor audience
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.
Applied AI Scientist
Probabilistic forecasting with a financial-impact framing is exactly the applied AI scientist role at any utility, trading, or grid-services company.
This challenge sharpens
- time-series-forecasting
- probabilistic-modeling
- model-evaluation
Machine Learning Engineer
Production-ready training + inference + monitoring spec is the MLE deliverable for any forecast that touches money.
This challenge sharpens
- ml-pipelines
- feature-engineering
- model-evaluation
Data Scientist
Walk-forward backtests, reliability diagrams, and impact translation are core data-scientist disciplines.
This challenge sharpens
- time-series-forecasting
- probabilistic-modeling
- feature-engineering