Vacancy "Statistics and A/B Testing Analyst at T-Bank" (Category HR)
I’ve been talking a lot lately about working with data, doing experiments, and building an a/b platform. And it's not for nothing - we create an a/b platform for the whole company, and we also have a laboratory of applied statistics, which is headed by Roman Filev. (@simbaizdolgopi). Roma is now looking for analysts to develop and implement SOTA hypotheses testing methods, pioneer best practices and roll them through a single decision-making platform. (A/B platform, product analytics platform, etc.). And these guys often parachute into food teams to help in difficult situations in this area.
In duty Analysts will be included Research and application of modern mathematical and statistical methods in A/B testing Communication with business units to accurately formalize business tasks Development and improvement of A/B testing methods Dissemination of best practices of A/B testing through participation in educational projects and presentations at specialized conferences
The following are necessary for successful performance of duties theory -> Probability theory and statistics: Strong possession of basic courses A thoughtful understanding of testing statistical hypotheses, including Type I and Type II errors and p-value interpretation Experience with basic statistical criteria, including Z/T/U tests and Kolmogorov-Smirnov, Anderson-Darling criteria Knowledge of limit theorems, including the law of large numbers (Zbc.)The Central Limit Theorem (CTT) and the Donsker-Prokhorov theorem -> A/B testing: Deep understanding of the statistical mechanics of A/B testing, with the ability to link tests to hypothesis testing Skills in determining sample sizes, working with multiple hypothesis testing and ratio-metrics Awareness of current approaches for accelerating A/B testing, such as reducing variance and consistently testing hypotheses -> Metric theory: Understanding the different types of metrics and the principles of their choice Acquaintance with approaches to creating a hierarchy of metrics
Minimum practical requirements Have a middle analytics level in SQL and Python Experience with click, gp, superset and visualization tools and project management systems (Jira/Confluence)
If you like the vacancy, write to Roma in person (@simbaizdolgopi) and get the details.
P.S. By the way, I have already talked about books on the topic of statistics that would be useful to such an engineer.
- How to Lie with Statistics (How to Lie with Statistics) - on the fingers explains how to lie with the help of statistics, and from here it becomes clear the motivation for creating a system of summing up the results of experiments
- Understanding Statistics and Experimental Design. How to Not Lie With Statistics (Statistics and Experiment Planning for the Uninitiated) This book talks about the design of experiments and the mathematics behind them.
- Confidential a/b testing (Trustworthy Online Controlled Experiments) This book also provides insights into how to build a platform for company-wide experimentation.
- Dark data (Dark Data. Why What We Don’t Know Is Even More Important Than What We Do) A book about how data can be screwed up and what can be done about it
#Vacancy #Statistics #Data #Analytics