Senior Data Scientist at New York Jets

The New York Jets are building a world-class football analytics program that turns complex data into clear, actionable decisions. As a Senior Data Scientist, you’ll lead high-impact modeling and research across player evaluation, coaching support, and football strategy—while setting the technical standard for how we design, validate, deploy, and communicate models at scale. This role is for a collaborative, high-ownership teammate who can architect end-to-end pipelines and evaluation processes, and who is fluent in modern deep learning within a competitive environment. Core Responsibilities: • Architect modeling systems that are repeatable, auditable, and production-ready: data → features → training → evaluation → delivery/monitoring. • Design evaluation frameworks (calibration, uncertainty estimation, bias/error analysis) to ensure models are ready to support high impact decisions. • Deliver research and analysis for: • Player evaluation (e.g., forecasting, role/fit analysis, contextualization across situations). • Coaching support requests (rapid-turn insights with clear assumptions/limitations). • Football research (longer-horizon studies that improve the organization’s decision-making). • Partner cross-functionally with coaching, scouting, performance, and football operations to define problems, translate questions into measurable outcomes, and communicate results clearly. • Mentor and raise the bar for DS best practices: code quality, review, documentation, experimentation standards, and reproducibility. Required Qualifications: Football / Domain - • Prior NFL or sports experience is preferred (club, league, or NFL-adjacent role where you shipped analytics that informed football decisions). • Strong understanding of football, with the ability to translate football questions into rigorous analyses without over-complicating the output. • Proven ability to collaborate in high-trust environments with diverse stakeholders and tight timelines. Technical Leadership & Modeling Architecture - • Demonstrated experience owning end-to-end modeling pipelines—including data sourcing, feature design, training, evaluation, deployment/packaging, and ongoing monitoring. • Expertise in building evaluation processes beyond a single model: • Principled baselines. • Back testing and leakage prevention. • Uncertainty quantification and calibration. • Clear decision recommendations and tradeoffs. • Ability to operate at “Principal” level: scoping ambiguous problems, setting technical direction, and aligning stakeholders. Deep Learning - • Hands-on experience applying deep learning in production or near-production settings, including on or more of the following: • Neural nets and/or transformers • Representation learning / embeddings. • Modern training practices (regularization, optimization, early stopping, hyperparameter search). • Proficiency with at least one deep learning framework (e.g., PyTorch or TensorFlow) and comfort working with GPU-enabled workflows. Data & Software Foundations - • Strong proficiency with Python or R and SQL. • Strong software engineering habits: • Version control (Git), code review, testing. • Modular, maintainable codebases. • Documentation and reproducibility (experiment tracking, model/version provenance). • Experience integrating data from multiple sources and producing reliable datasets for downstream modeling and reporting. Communication & Teamwork - • Excellent written and verbal communication—able to explain methods, results, and limitations to both technical and non-technical partners. • Demonstrated ability to deliver both: • Rapid-turn answers (triage, directional insight). • Deep research (rigorous studies with clear takeaways). • Low-ego teammate: collaborative, pragmatic, and committed to shared success. Preferred Qualifications: • Master’s degree or higher in a quantitative field (e.g., Statistics, CS, Applied Math, Physics, Engineering, Economics). • Experience with one or more of: • Causal inference. • Bayesian/hierarchical modeling. • Time-series or survival modeling • MLOps tooling (containers, orchestration, CI/CD patterns, model registries). • Databricks or similar orchestration platforms. • Experience leading and mentoring other data scientists and influencing technical standards across a team. ...