Spectral Clustering for an Urban-Mobility Operator's Network
Overview
What this challenge is about.
Build a weighted O-D graph from 4M trips, run spectral clustering on a mobility network, and recommend optimal zones. Get a verifiable certificate.
The scenario
The scale-up (around 110 people, around 15,000 active vehicles across Mexico City, Bogota, and Lima) spends around USD 1.2 million per year on rebalancing trips; even a 10% reduction is meaningful in a contribution-margin-sensitive business.
The Brief
What you'll do, and what you'll demonstrate.
Use spectral clustering on the O-D trip graph to redesign service zones for a shared-mobility fleet and quantify rebalancing-cost savings.
Earning criteria — what you'll demonstrate
- Construct weighted O-D graphs from trip data
- Apply spectral clustering using the graph Laplacian
- Evaluate clustering choices against operational KPIs
- Communicate algorithmic zone redesign to an operations team
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Machine Learning on Graphs
Master · Machine Learning
Strong alignment
This challenge maps to Machine Learning on Graphs at the Master level. It sharpens the same practical skills your coursework expects — but in a real industry context with actual constraints and deliverables.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Spectral Methods
Apply spectral methods to solve real industry problems and demonstrate production-level capability.
- Spectral Clustering
Apply spectral clustering to solve real industry problems and demonstrate production-level capability.
- Graph Laplacian
Apply graph laplacian to solve real industry problems and demonstrate production-level capability.
- Graph Construction
Apply graph construction to solve real industry problems and demonstrate production-level capability.
- Evaluation
Apply evaluation to solve real industry problems and demonstrate production-level capability.
- Python
Write clean, efficient Python for data processing, automation, and backend services.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Data Scientist
Applying spectral methods to a real operational graph and delivering a zone map operations can use is exactly the day-one job of a data scientist at any mobility or logistics company.
This challenge sharpens
- spectral-methods
- spectral-clustering
- evaluation
Applied AI Scientist
Translating algorithmic clustering into an operations-ready GeoJSON deliverable is core applied-AI-scientist work in mobility analytics.
This challenge sharpens
- spectral-clustering
- graph-construction
- graph-laplacian
Data Engineer
Building reproducible graph-construction + geospatial pipelines transfers to data-engineering roles on any urban-mobility data team.
This challenge sharpens
- graph-construction
- python
- graph-laplacian