Build a Search-Based Logistics Router for a Same-Day Delivery Service
Overview
What this challenge is about.
Build a Python service exposing /route that takes (courier_location, list_of_stops, traffic_window) and returns an ordered route. Implement 2 search strategies: A* (with a domain heuristic such as straight-line distance with traffic penalty) and a comparison baseline (greedy nearest-stop, or simulated annealing). Run on 2 weeks of anonymized real orders (provided). Evaluate against hand-planned baselines on: total distance, on-time-delivery rate, planning latency. Build a small visualization (Streamlit) so the COO can spot-check routes. Deliver: GitHub repo with both strategies + tests, Streamlit dashboard, 5-page evaluation report, and a deployment recommendation (which strategy to ship).
The Brief
What you'll do, and what you'll demonstrate.
Build a classical-AI search-based router that beats hand-planned baselines on distance + on-time rate, with a 2-second planning latency.
Earning criteria — what you'll demonstrate
- Apply A* search to a real combinatorial routing problem
- Design a domain-specific heuristic and measure its admissibility/effectiveness
- Evaluate AI algorithms against a real human-decision baseline
- Translate algorithm choice into a defensible engineering recommendation
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.
Product Manager
PMs who can evaluate AI vs. human baselines defensibly avoid the classic 'AI replaces dispatcher' overclaim that erodes operator trust.
This challenge sharpens
- algorithm-evaluation
- heuristic-design
- intelligent-agents