Sr Data Engineer – Sports Analytics Platform (Python, Web Scraping, Statistical Modeling) - FT - Worldwide

About the project

An NCAA Division I women's volleyball program is building a private player-evaluation platform for roster, recruiting and transfer-portal decisions. The core idea: rate players on efficient, high-value production, not the stats that look best in highlights.

The coaching staff has explored the concept but has no production code, so this is a ground-up build. You'll work directly with the head coach and the team's analytics lead, who know exactly what they want to see.

This is a full-time, focused engagement. We're looking for someone who can commit 40 hours a week until it's delivered, not someone splitting time across projects.

Phase 1: Core platform

• Scrapers for season stats from about 340 Division I athletics sites. The sites run on multiple platforms, some rendered in JavaScript. Data refreshes at least daily in season.
• Player matching: one record per player across seasons, schools, transfers and name variations, joined to roster data (position, class, height).
• A PostgreSQL database of players, teams, conferences and seasons that supports historical comparison.
• A position-specific value rating (points added per set) with conference-strength adjustments. Staff can change weights without code changes.
• A secure, mobile-friendly dashboard: rankings, filters, player rating breakdowns, weight controls, and a chart showing undervalued players.
• Deployment on client-owned cloud accounts, with documentation and a maintenance guide.

Phase 2: Match data

• Parse licensed VolleyMetrics / DataVolley match files.
• Build an expected-value model that values every contact by its effect on rally win probability.
• Merge the results into the dashboard. The data stays on the program's systems.

Phase 3 (optional, may be a separate hire): Practice movement tracking

• Camera-based tracking of player movement, such as blocker closing speed and transition timing.
• Computer vision experience is a bonus, not a requirement.

Must-have

• 5+ years of experience; strong Python (pandas, NumPy)
• Production web scraping: requests/httpx, BeautifulSoup/lxml, Playwright or Selenium, including maintaining scrapers when sites change
• Entity resolution / record linkage on messy data
• PostgreSQL and scheduled pipelines (cron, GitHub Actions, Prefect or Airflow)
• Solid statistics and probability (statsmodels or scikit-learn)
• A dashboard framework (Streamlit, Dash or React) with authentication
• AWS, GCP or Azure deployment; Git/GitHub
• Able to own a build end to end and communicate clearly with non-technical stakeholders

Strong plus

• Sports analytics or sports betting experience, especially win-probability or expected-value models
• Volleyball knowledge, DataVolley/VolleyMetrics, or R
• Computer vision / pose tracking (for Phase 3)
• University SSO integration

Why this role

• Direct work with a Division I coaching staff on a product they'll use every season
• Clear path to ongoing work: maintenance, new data sources and features, with potential for a long-term retainer as the platform grows