Data Scientist, Personalization and Recommendation
We are seeking data scientists who have a solid background in machine learning and have strong track record of implementing real world user-facing applications. The candidate should be well-versed in data analysis and algorithm implementation, ready to be let loose on Tumblr’s petabytes of data.
What you’ll do:
Apply data mining and machine learning techniques to develop better personalization, recommendation, and content discovery.
Define metrics, conduct A/B testing, and oversee statistical measurement of new algorithms and approaches.
Design and develop effective models, features, and algorithms involving user activities and interests, blogs and posts, social graphs, etc.
Use MapReduce frameworks to generate production data for recommendation and discovery features.
What we’re looking for:
A strong passion for empirical research with massive data.
Experience of solving real problems with data mining and machine learning techniques.
Experience of large datasets analysis with tech stack such as Hive, Scalding, Spark.
Familiarity with Linux-based systems.
Strong computer science background preferred.
Tools We Like:
People come to Tumblr to expand their perspectives. They join a community and a cultural dialogue that exposes them up to new ideas, new passions, and new forms of self-expression. People seem to really like this thing we’ve built, and we need smart, humble, funny people to keep it running and growing. We’re big enough to give you the resources you need and still small enough that you can work on things that matter. Come work with us.