Sr. Data Scientist, Recommendations Los Angeles, California Department Product Analytics

Launched in 2012, Tinder® revolutionized how people meet, growing from 1 match to one billion matches in just two years. This rapid growth demonstrates its ability to fulfill a fundamental human need: real connection. Today, the app has been downloaded over 630 million times, leading to over 97 billion matches, serving approximately 50 million users per month in 190 countries and 45+ languages - a scale unmatched by any other app in the category.

In 2024, Tinder won four Effie Awards for its first-ever global brand campaign, "It Starts with a Swipe"™

The Data Science & Analytics team thrives on data-driven insights to make more informed decisions through our insights into our member’s behavior, preferences, and common trends. We take ownership over the integrity of our data and work to improve data literacy across Tinder. Recommendations (Recs) is core to Tinder’s experience—covering ranking, retrieval, signals, and model evaluation—to improve match quality, conversations, retention, and revenue through principled ML and experimentation.

As a Senior Data Scientist on the Recommendations (Recs) team, you will partner closely with Product, Engineering, and Machine Learning (ML) to identify and size new opportunities, strengthen existing algorithms, and shape measurement and experimentation across a two-sided marketplace. You’ll build the tooling and dashboards needed for crisp reads and health monitoring, communicate insights and tradeoffs with executive clarity, and serve as a trusted, respected partner to the Recs pod and a mentor for the broader Data Science team.

This role will be given wide latitude but high expectations for the development of analyses, models, and plans that will be shared directly with the executive team and help shape the trajectory of our product roadmap.

This is a hybrid role and requires in-office collaboration three times per week in Los Angeles, Palo Alto or San Francisco.

• Work closely with Product, Engineering, and ML to identify and evaluate new opportunities; frame hypotheses, define success metrics and guardrails, and translate findings into clear product recommendations.
• Support the ML team in improving algorithms across retrieval, ranking, and personalization; strengthen offline/online evaluation and alignment.
• Define and lead experimentation design and analysis tailored to a two‑sided marketplace; drive meta-analyses and playbooks that uplevel reads and decision quality.
• Build tools and dashboards to improve experiment reads and KPI monitoring; standardize templates and health checks for fast, reliable iteration.
• Deliver executive-ready presentations and docs that clarify options, tradeoffs, risks, and expected business impact.
• Be a trusted and respected partner for the Recs pod, focused on delivering the best recommendations for our worldwide member base.
• Mentor and inspire other data scientists; review analyses and elevate experimentation, causal inference, and model evaluation practices across the team.

• Bachelor’s, Master’s, and/or Ph.D. degree in a quantitative field (e.g., Statistics, Mathematics, Computer Science, Economics, or related fields).
• 5+ years of professional experience in data science/analytics, with meaningful time in recommender systems, ranking, search, or personalization at consumer scale (or equivalent impact/complexity).
• Fluency in SQL and Python (required).
• Deep understanding of statistics and causal inference; hands‑on experience designing and analyzing online experiments (A/B, variance reduction, sequential testing) and applying quasi‑experimental methods when appropriate.
• Strong product sense and analytical rigor; ability to frame the right questions, choose fit‑for‑purpose methods, and land actionable recommendations with cross‑functional partners.
• Familiarity with machine learning for recommendations, including offline/online metric design and model evaluation for ranking/personalization use cases.

• Experience with modern Recs stacks (e.g., retrieval/two‑tower, learning‑to‑rank, embeddings/feature stores) and counterfactual evaluation approaches (IPS/DR, switchback…