Strava Lead Data Scientist at Strava responsible for defining evaluation frameworks and measurement strategies for ML products. Drive model quality and identify AI/ML product opportunities.
Responsibilities
The Data Science team at Strava works across the organization to find solutions to the highest-leverage, and often the most challenging, problems facing the business. We use machine learning, causal inference, and measurement systems to synthesize Strava’s unique data assets into models, metrics, and recommendations that our leadership team can act on with confidence.
We are looking for a Data Scientist to join the Data Products team at Strava, a team at the core of Strava’s AI strategy, responsible for turning Strava's unique community and activity data into reliable, reusable, enriched datasets powering user experiences at scale. This is a strategic individual contributor role that will partner closely with a growing team of Machine Learning Engineers. You'll drive the team’s measurement strategy and define the development feedback loop— defining evaluation standards, surfacing where performance is breaking down, and steering the team toward the highest-impact opportunities.
We follow a flexible hybrid model that translates to more than half of your time on-site in our San Francisco office — three days per week.
Define what "good" looks like for Strava's internal models and the ML products built on top of it, setting the evaluation frameworks, offline and online metrics, and quality bars the team steers by.
Build the measurement layer for cross-domain ML products, scaling the org’s visibility into performance across the business
Own experimentation and monitoring for production models and contribute to monitoring of key metrics and drift detection for enriched datasets, ensuring quality for downstream athlete-facing experiences.
Identify and size AI/ML product opportunities, translating ambiguous problem spaces into scoped initiatives with defined success criteria.
Serve as the data science domain expert for the team, raising the standard for baselines, validation, and evidence quality across ML engineers and cross-functional partners.
What You’ll Bring to the Team:
5+ years of experience in data science or a related quantitative domain, including hands-on ownership of model evaluation and measurement for systems running in production.
Experience defining evaluation frameworks for ambiguous ML problems, including baseline selection, offline and online metric design, and validation strategies where ground truth is imperfect
Strong SQL proficiency and comfort writing production-quality Python for statistical data processing.
Qualification
Strong SQL proficiency
Required
5+ years of experience in data science or related quantitative domain
Experience owning measurement strategies
Experience employing experimental and quasi-experimental methods
Depth in causal inference methods and their real-world failure modes
Strong SQL proficiency
Comfort writing Python for statistical data processing
