Basketball Data Scientist

About the job Company Description

Wisely Optimize is a basketball analytics company with 11 clients across the country — and growing fast. We turn complex data into simple, actionable workflows for coaches and decision-makers, especially around roster construction and the transfer portal window. We're in a real growth moment: our client base is expanding, our user base keeps climbing, and inbound interest is building week over week. You'd be coming on as the momentum picks up. Learn more at wiselyoptimize.com.

You'd be building alongside Co-Founder & CEO Alex Beene, who has worked with 5 pro basketball teams across coaching and front office, and Basketball Software Engineer David Simmerman, a decade-plus engineer (ex-Director of Frontend, XGen AI) and lifelong basketball fanatic who builds analytics tools for the love of the game. This is an equity-only opportunity for someone who wants to build long-term with a growing company, make a real impact on the basketball world, and have a career in Basketball Analytics.

Role Description

We're looking for a data scientist to lead our modeling work day to day — the player evaluation, projections, and fit analysis that coaches use to make roster decisions — and to get that work into production where it actually reaches users. Some models will be yours from scratch. Others you'll inherit from Alex, pressure-test, sharpen, and productionalize. Alex still builds models himself and will coach you on the basketball side, so this works best for someone who isn't precious about whose model it is and cares more about what ships and gets used.

The work spans the full path from idea to impact: feature engineering, validation, and knowing when a simple explainable model beats a complex one; taking models from notebook to production with clean code and reproducible pipelines; and doing the data engineering it takes to keep them fed with fresh data. Just as important is translating output into numbers a staff understands and acts on, rather than a black box they ignore. You'll bring real analytical rigor — sample size, noise, role dependence, pace and opponent context — and say plainly where the numbers lie.

This is equity-only and designed to fit around a full-time job: remote, async, and outcome-driven, so you can build something world-class without quitting your day job right away. Outcomes matter more than hours, though we do keep an afternoon daily standup rhythm and expect clear communication about progress and blockers.

Qualifications

Proven, elite analytical work. This is the one area we need you strong in from day one — it's not learn-on-the-job. You're excellent with statistical modeling, validation, and knowing what a number actually means. Strong Python and SQL, and comfortable working directly with data rather than waiting for someone to hand you a clean table. Comfortable with engineering, not just notebooks: you can write code others can run, get a model into production, and do the data engineering work it takes to keep it running. Deep knowledge of advanced basketball analytics — you know the modern metrics, their tradeoffs, and can argue about how to measure a player with numbers. Coachable: happy to inherit someone else's model, take feedback, and get coached on the basketball side. What ships matters more than whose idea it was. Discernment over complexity: you'd rather ship a model a coach trusts than a model that scores well and never gets used. High motor and follow-through: you communicate clearly and ship consistently. A passion for basketball and a desire to work in Basketball Operations long-term. Nice to have: experience with basketball data sources and their nuances; work that's been used to make a real decision (team, front office, or published research); B2B SaaS or analytics product experience; experience in fast-moving, small-team environments.

How to apply

Apply on LinkedIn, then email [email protected] with:

Your background (1–2 paragraphs) Links that show your analytical work (GitHub, projects, research) Your answer to this, in your own words: What single number would you use to evaluate how good a college basketball player is? What is flawed about that number, and why did you choose it? Explain your formula and your reasoning.

An answer to the last question is required; applications without one won't be considered.