Kirill Khrylchenko
Research Scientist, Senior Lecturer at HSE University
Moscow, Moscow, Russia
Summary
👤
Senior🎓
Top SchoolKirill Khrylchenko is a research-focused machine learning engineer and lecturer with 10 years of experience specializing in recommender systems, deep learning and large-scale personalization. He led R&D and production deployments of transformer- and graph-based ranking/retrieval models across major Yandex products, built the core DL personalization framework used at scale, and authored papers on logQ correction and scaling recommender transformers. At Yandex he combined hands-on research, architecture and people leadership—growing a 10–13 person team, running the largest SAR reading group, and presenting 40+ papers—while earlier work tackled petabyte-scale data pipelines and GPU batching for billions of user embeddings. Now teaching and researching sequential recommendation, generative retrieval and contrastive learning at HSE and YSDA, he blends industrial impact with academic rigor and a knack for turning research prototypes into high‑QPS production systems. An unexpected thread through his career is a deep grounding in probabilistic models from his topic-modeling research, which informs his practical approaches to generative and contrastive methods in modern RecSys.
10 years of coding experience
6 years of employment as a software developer
Master's degree Mathematics and Computer Science, Master's degree Mathematics and Computer Science at Moscow State University
Github Skills (2)
dm10
eris10
Programming languages (2)
Github contributions (5)
ziz11/CEV-Eris
Mar 2019 - Mar 2019
Death is our destination
Contributions:2 PRs, 1 push in 1 day
destinationdeath
ziz11/ultra_secret_scripts
Mar 2019 - Mar 2019
This is ultimately secret what can I say...
Contributions:2 PRs, 1 push in 1 day
saysecret
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.