Hi! My name is Vladimir Kukushkin, and I have over 13 years of experience as a data scientist / data analyst. The slash signifies that I typically work at the intersection of product analytics and data science, tackling problems that require advanced techniques like predictive analytics, causal inference, and building statistical models. My specialization is user behavior analysis a.k.a. quantitative UX research or, more broadly, sequential data analysis. Lately I’ve also been turning this lens on AI products themselves, looking at how users, and increasingly agents, actually behave inside LLM-based systems.
My projects:
- retentioneering. An open-source Python library for user behavior analysis: I’m a co-founder and maintainer. In 2026 I led its ground-up rewrite into version 5.0: a new DuckDB-based engine, interactive Jupyter widgets, and an MCP server that lets LLM agents run behavioral analyses on their own.
My publications:
- Vladimir Kukushkin - Modeling DAU with Markov Chain (TowardsDataScience Editors’ Pick)
- Anna Lioznova, Alexey Drutsa, Vladimir Kukushkin, Anastasia Bezzubtseva - Prediction of Hourly Earnings and Completion Time on a Crowdsourcing Platform. KDD-2020
- Olga Megorskaya, Vladimir Kukushkin, Pavel Serdyukov - On the Relation Between Assessor’s Agreement and Accuracy in Gamified Relevance Assessment. SIGIR-2015
- Anna Lioznova, Anastasia Bezzubtseva, Alexey Drutsa, Vladimir Kukushkin - Method and system for determining productivity rate of user in computer-implemented crowd-sourced environment. US Patent 11475387
Books I recommend: