Senior Data Scientist, Twitch

• BA/BS in Analytics, Data Science, Computer Science, Mathematics, or equivalent industry experience
• 5+ years of experience as a data scientist or data analyst in a high velocity, data-driven environment
• 5+ years of machine learning/statistical modeling data analysis tools and techniques, and parameters that affect their performance experience
• Expert SQL skills and proficiency in Python/R
• Experience using data to create insight, drive business decisions and influence leadership

Desirable

• The Community Discovery Machine Learning (CDML) team powers the systems that help millions of active viewers find the creators and communities they'll enjoy most. Powering the home page recommendations, browse, search, and feed ranking experiences, this team and its algorithms are at the heart of enabling the connections that make Twitch a great place for so many users
• Data is critical to the algorithms that power the recommendation, search, and ranking systems. It's also critical to making decisions, especially working on systems that are themselves data-driven. As a Senior Data Scientist on the CDML team, you'll be responsible for helping drive improvements to the machine learning systems as well as analytics to drive decision-making
• While there is a team of Applied Scientists building and shipping the algorithms themselves, data science can help improve these systems directly. In this role, you can identify and build new signals to input into the models. We're also working on the value model that the algorithm optimizes, and your input will be critical to understanding the tradeoffs and balancing multiple objectives in a scientific way
• We also still have big unanswered analytics questions to solve. How often do viewers just want to get to the content they already know they want to watch, and when are they open to exploring new channels? These are the sorts of questions you'll be tackling
• You can work in San Francisco, CA; New York, NY; or Seattle, WA
• Inform product strategies by defining and updating core metrics for each initiative
• Estimate the opportunity sizing of new features the team could take on
• Identify and build new signals to incorporate into the algorithms driving recommendations, search, and feed ranking at Twitch
• Identify metric tradeoff ratios that help inform value model choices, long-term impact from early-growth-funnel users, and other product decisions
• Establish analytical framework for your team: ad-hoc analysis, automated dashboards, and self-service reporting tools to surface key data to stakeholders
• Design A/B experiments to drive product direction with iterative innovation and measurement
• Work hand-in-hand with business, product, engineering, and design to proactively influence and inform teammates' decisions throughout the product life cycle
• Distill ambiguous product or business questions, find clever ways to answer them, and to quantify the uncertainty