Profile

Experienced data scientist. Enjoy digging into data and building statistical models to solve business problems.
  • Extensive knowledge of Bayesian statistics and solid work experience with a variety of ML algorithms for classification and regression (logistic regression, decision trees, boosting, SVMs, etc.).
  • Intimately familiar with the entire ML workflow, from data gathering/cleaning/exploration to feature engineering to model verification and deployment. Experienced in natural language processing techniques.
  • Able to code in R, Python, C++, Java. Good working knowledge of Hadoop and Map/Reduce.
  • Comfortable with making decisions under uncertainty. Worked as individual contributor, but also managed and led a team of engineers and scientists.
  • Excellent communication skills; able to clearly explain complex problems and solutions.
Currently a data scientist at eBay Inc. as part of my ongoing PhD work at Applied Mathematics and Statistics Department in the School of Engineering at UC Santa Cruz. My research focuses on methods of causal inference in observational studies.

Previously, spent a number of years as data scientist / research engineer at Yahoo! (Search and Labs), working on a number of machine learning problems in the area of natural language processing and advertising. Prior to that, founded and served as CTO of Toolwire Inc.. I enjoy teaching, and have served as a part-time faculty mathematics instructor at De Anza Community College.

I hold a MS degree in Statistics and Applied Mathematics and a MS in Computer Science, both from the University of California, Santa Cruz. Also, I hold BS degree in Electrical Engineering and Computer Science from the University of California, Berkeley.

I can be reached at : vadim _at_ numberjack.net