Overview
What this challenge is about.
Encode the dispatch problem (which customers to curtail by how much, respecting per-customer contractual caps and grid-cell totals) as a SAT or MaxSAT instance. Solve 50 historical events with an off-the-shelf solver under a 60-second time budget. Compare against the greedy heuristic on objective value (curtailment delivered) and feasibility. Deliver a 3-page memo + an architecture sketch of what production integration would look like.
The Brief
What you'll do, and what you'll demonstrate.
Decide whether a SAT/MaxSAT planner beats the greedy heuristic on curtailment delivered within a 60-second dispatch budget.
Earning criteria — what you'll demonstrate
- Encode a real planning problem as SAT/MaxSAT
- Use an off-the-shelf solver under a strict time budget
- Compare against an industry baseline with realistic event data
- Reason about production integration of a SAT-based component
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.
ML Researcher
Rigorous benchmarking of a SAT-based method against a production heuristic is exactly the experimental discipline ML researchers practice.
This challenge sharpens
- sat-based-planning
- benchmarking
- experiment-design
AI Engineer
SAT encoding on a real ops problem with a production-integration sketch is high-value AI engineering work in grid + supply-chain startups.
This challenge sharpens
- constraint-encoding
- domain-modeling
- sat-based-planning
Applied AI Scientist
Turning a solver experiment into a memo with revenue framing is what applied AI scientists do at infrastructure-AI companies.
This challenge sharpens
- benchmarking
- experiment-design
- domain-modeling