Ilia Gerasimov
Software Engineer at Google DeepMind
London, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolIlia Gerasimov is a seasoned software engineer with nine years of experience building and leading storage and distributed systems, currently driving storage infrastructure for Gemini training and serving at DeepMind in London. Previously he led LogDevice development at Meta, moving exabytes of data with C++/Rust at extreme scale and improving reliability through work on metadata trimming, API handling, and connection/session refactors. Earlier roles include leading backend teams at Yandex and building low-latency mobile and server systems, giving him deep expertise across Scala/Akka, Cassandra, JVM and high-performance C++. He combines hands-on systems design and performance tuning with people leadership—hiring, mentoring and long-term architecture—and is comfortable working on the gritty plumbing that keeps ML and data platforms running. A not-obvious strength: he has a track record of improving operational robustness by instrumenting failure modes and dynamically tuning background maintenance tasks.
9 years of coding experience
17 years of employment as a software developer
Bachelor of Applied Mathematics and Computer Science, Bachelor of Applied Mathematics and Computer Science at Regional Cisco network Academy in Yaroslavl State UniversityBachelor Computer Science, Bachelor Computer Science at Yaroslavl Demidov State University
Russian, English
Github Skills (9)
c-language10
thrift10
cprogramming-language10
networking9
multithreading9
logging8
memory-management8
performance-optimization8
system-design8
Programming languages (1)
Github contributions (5)
facebookarchive/LogDevice
Feb 2020 - Oct 2020
Distributed storage for sequential data
Role in this project:
Back-end Developer
Back-end DeveloperContributions:70 commits, 3 comments in 8 months
Contributions summary:Ilia primarily contributed to the `logdevice` repository by implementing and modifying features related to metadata log trimming and the handling of Thrift API calls. They implemented new features, such as dynamically changing the run interval for metadata trimming and added counters for failures in writing metadata logs. They also refactored code related to connection management, session handling, and general API management. These changes were focused on improving performance, reliability, and the underlying architecture of the LogDevice server.
distributed-storagedatabasestoragesequentialsequential-data
747mmHg/747mmHg.github.io
Mar 2017 - Dec 2024
Contributions:2 PRs, 18 pushes, 2 branches in 7 years 10 months
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.