The Position
You will be a key member of the Interests Data Science team working closely with data scientists, engineers, and product managers to improve major new products like Topics and Lists Recommendation. You will design and analyze experiments, propose metrics to summarize and improve the performance of recommender systems, and help drive a deeper, quantitative understanding of interests and communities on Twitter.
Why it matters
Making it easier for users to find content related to their interests is core to Twitter’s strategy. The Interests data science team is on the leading edge of this work, and our analyses help motivate, design, and evaluate major product investments in this space.
What you’ll be doing
As a staff data scientist on the Interests team you will:
- Design, implement, and analyze experiments to determine how changes to the platform affect user behavior and refine our theoretical understanding of users’ interests and the communities they form.
- Conduct analyses to better understand the performance of our content recommendation, ranking, and clustering systems. Use this understanding to evaluate the performance of these systems and suggest improvements to them.
- Create accurate and statistically sound summary metrics to evaluate the performance of Interests related products and motivate teams’ targets/KRs.
- Communicate findings to executives and cross-functional teams.
Qualifications
What it takes
- Advanced degree in a discipline that uses statistical analysis and 4+ years of experience (or 7+ years of total experience).
- Experience with one or more of the following in an applied business setting: advanced statistical techniques for A/B testing, methods for experimental design, observational causal inference, or quasi-experimental analysis.
- Solid understanding of the design and analysis of recommender systems.
- Track record for executing independent projects and leading complex, multi-functional projects with several dependencies.
- Strong proficiency with Python / R and SQL.
- Experience mentoring junior data scientists and enthusiasm for building out a new data science team.