Data Scientist - Statistical Inference
- San Francisco, CA, USA
Optimizely is the world's leader in customer experience optimization, allowing businesses to dramatically drive up the value of their digital products, commerce and campaigns through its best in class experimentation software platform. By replacing digital guesswork with evidence-based results, Optimizely enables product and marketing professionals to accelerate innovation, lower the risk of new features, and drive up the return on investment from digital by up to 10X. Over 26 of the Fortune 100 companies choose Optimizely to power their global digital experiences. Optimizely’s impressive customer list includes eBay, FOX, IBM, The New York Times and many more global enterprises.
As a Data Scientist at Optimizely, you’ll be in charge of Stats Engine (https://www.optimizely.com/products/statistics/), the core statistics methodology behind all our experiments. You will be a member of the Machine Learning/Statistics team and work on the cutting edge of statistics, helping answer questions like:
How can we safely employ multi-armed bandits in an A/B testing context? (https://medium.com/engineers-optimizely/stats-accelerator-acceleration-under-time-varying-signals-61ae857a436d)
Are approximate counting methods compatible with conventional hypothesis testing methods? (https://medium.com/engineers-optimizely/approximate-counting-and-statistical-significance-two-great-ideas-that-dont-play-nice-2bd643287644)
You’ll work with our product and engineering teams to build new statistics features for all of Optimizely’s products. You will get on customer calls to articulate how we maintain accuracy in our products as well as triage questions around statistical inference. You will educate internal and external stakeholders about core statistics concepts via webinars, speaking engagements, blog posts, and white papers. You will conduct original research and publish results in academic and industry conferences.
PhD in Statistics or a closely related field, or 5+ years of equivalent industry experience in A/B testing and digital experimentation
Strong research interest and experience with statistical inference. Especially: design of experiments, randomized controlled trials, sequential testing, robustness, and bayesian inference.
Experience with Python and/or Java preferred.
Some experience in math education such as teaching, teaching assistantships, consulting, or conference talks.
Some of our public work:
Stats Engine White Paper - (http://pages.optimizely.com/rs/optimizely/images/stats_engine_technical_paper.pdf)
Stats Engine Academic Paper - (https://arxiv.org/abs/1512.04922)
Bayesian vs Frequentist Statistics (https://blog.optimizely.com/2015/03/04/bayesian-vs-frequentist-statistics/)
Optimizely’s Stats Engine (https://blog.optimizely.com/2015/01/20/statistics-for-the-internet-age-the-story-behind-optimizelys-new-stats-engine/)
Approximate Counting and Statistical Significance (https://medium.com/engineers-optimizely/approximate-counting-and-statistical-significance-two-great-ideas-that-dont-play-nice-2bd643287644#.6uyxiytlr)
CODE @ MIT 2016: A/B Testing in a Changing World
INFORMS 2015: Can I Take a Peek? Continuous Monitoring of A/B Tests
CODE @ MIT 2014: Can I Take a Peek? Continuous Monitoring of A/B Tests
At Optimizely, we embody inclusion and embrace diversity. Optimizely is an equal opportunity employer and makes employment decisions on the basis of merit. Optimizely prohibits discrimination based on race, color, religion, sex, sexual identity, gender identity, marital status, veteran status, nationality, citizenship, age, disability, medical condition, pregnancy, or any other unlawful consideration. All your information will be kept confidential according to EEO guidelines.