I worked on a college basketball coaching analytics MVP designed to support coaches with game prep, scouting, player and team analytics, play management, and in-game decision support.
The product needed to calculate useful information on the fly, which meant the backend had to do a lot of work before the user experience could become meaningful. It needed data collection, schemas, APIs, calculations, documentation, and a structure that another team could build on.
I designed the backend architecture and built MongoDB schemas with Mongoose to support the product’s data model. I also wrote Node-based web scrapers to collect and parse public sports data, including team stats, player stats, game stats, play-by-play data, schedules, rosters, box scores, rankings, and other information needed to power real-time calculations.
The challenge was not just gathering data. The data had to become usable for coaching workflows. That meant thinking through how different data points related to one another, how calculations should be supported, how APIs should expose the right information, and how the frontend team would eventually turn that backend foundation into a usable coaching experience.
I developed APIs to manage users, services, calculations, and data points. I also documented the backend system so the product could be handed off more cleanly and the next team could understand how the pieces fit together.
Once the backend foundation was in place, I helped hire and grow the frontend team that would build out the application experience. I guided them through the architecture, helped translate the technical foundation into product workflows, and supported the transition from backend capability to usable MVP.
The MVP handoff went well because the system had enough structure to support the next stage of work: data sourcing, schemas, APIs, calculations, documentation, and a team that understood what had been built.
This project is a good example of how analytics products need more than charts. A useful analytics platform depends on data collection, data modeling, calculation logic, workflow context, and a technical foundation that can support real decisions.
