Java Programming, Python Programming, SQL, R Programming
Launched in 2011, Twitch is a global community that comes together each day to create multiplayer entertainment: unique, live, unpredictable experiences created by the interactions of millions. We bring the joy of co-op to everything, from casual gaming to world-class esports to anime marathons, music, and art streams. Twitch also hosts TwitchCon, where we bring everyone together to celebrate and grow their personal interests and passions. We're always live at Twitch.
About the Role:
Are you passionate about making the internet safer, more inclusive, and more enjoyable for people around the world? As a Data Scientist on the Community Health team at Twitch, you'll play a central role in the design and deployment of products that support the safety of millions of Twitch users. You will be part of a growing organization that builds the tools which help communities self-moderate, improves our reporting and enforcement processes, and builds automated machine learning models that serve as the first line of defense against harmful behaviors.
You will specifically lead research in areas such as (1) quantifying the prevalence of harmful actions across the service and (2) estimating the impact of new features on our viewers and creators, often in scenarios where traditional randomized controlled experiments are not possible. You will design research projects that help build our knowledge of how to keep communities safe, and leverage this expertise to guide strategy around the company. You will report to the Head of Science for Community Health. We are based in San Francisco, however we're open to you working out of our Seattle, WA, Irvine, CA, or Salt Lake City, UT offices.
You'll find a Requirements section below. If you meet all of these, then we encourage you to apply. If you meet most of them, and feel that you have a unique perspective or skill-set that can help us to design safer and more rewarding communities, then we encourage you to apply.
- Develop measures of the health and safety of our communities, estimate the prevalence of policy violations across the service, and measure the performance of product and policy programs designed to make Twitch safer.
- Apply causal inference approaches to estimate the impacts of both harmful actions and safety products, using observational data, using strategies that minimize bias.
- Build expertise in the design of product features and programs to support safer and more rewarding interactions within online communities; maintain awareness of recent publications, industry trends, and best practices that can be integrated into product development and strategy at Twitch.
- Scope and prioritize team needs; distill ambiguous product or strategy questions into testable hypotheses; find clever ways to test these hypotheses, estimate quantities of interest, and measure the uncertainty around these estimates.
- Maintain a culture of high-quality output and engagement with team members; communicate technical information to technical and non-technical partners; manage ad hoc requests and unexpected obstacles.
- Ph.D. in Computer Science, Economics, Statistics, Engineering, Mathematics, Data Science, or related field or MS/MA, with equivalent experience.
- Experience with advanced statistics and ML methods; expertise in causal inference methods such as instrumental variables, difference-in-difference methods, panel data models, regression discontinuity, and propensity score matching.
- Experience working with and manipulating large datasets; expert SQL skills and proficiency in Python or R.
- Comfort working on occasion with potentially sensitive or disturbing content.
- Natural curiosity about how to design successful online communities and eagerness to contribute to products that shape the community experience for millions of viewers and creators.
- 1+ years of experience applying causal inference, statistics, and ML methods to problems in the consumer internet or gaming space, or in a related context.
- Software development experience and familiarity with AWS or other cloud-computing services.
- Experience and interest in contributing to scientific publications in relevant, top-tier venues.
- Expertise in any of the following areas: NLP (e.g. topic modeling, sentiment analysis), social network analysis, fairness and bias assessment.
- Experience working on a Trust & Safety or Community Health team or conducting research on topics related to online harm.
- Medical, Dental, Vision & Disability Insurance
- 401(k), Maternity & Parental Leave
- Flexible PTO
- Commuter Benefits
- Amazon Employee Discount
- Monthly Contribution & Discounts for Wellness Related Activities & Programs (e.g., gym memberships, off-site massages),
- Breakfast, Lunch & Dinner Served Daily
- Free Snacks & Beverages
Twitch is a one-of-a-kind collision of subcultures, games, communities, and streamers co-creating content around everything they’re into.
This is more than just watching; this is multiplayer entertainment — unique, live, unpredictable, never-to-be-repeated experiences created by the interactions of many. Twitch brings the joy of co-op to everything gamers love, from anime marathons to esports competitions and more.
No. of Employees
San Francisco, CA, USA