Mark Andreev
Senior Software Engineer at Man Group
London, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolMark Andreev is a Senior Software Engineer based in London with 13 years of experience building high-performance, production-grade systems across fintech and industrial data platforms. He specializes in low-level performance optimization and distributed data architectures—having led Kafka + ClickHouse feature store migrations that yielded up to 15x query speedups and implemented Java 21 virtual-thread optimizations to cut CPU load by 5x. His background spans ML engineering and backend work, contributing bug fixes and new distance metrics to Apache Ignite and performance-focused refactors to projects like facebookarchive/nailgun. Comfortable across Java, Python, Kubernetes, and cloud/on‑prem deployments, he pairs strong algorithmic foundations from an MSU applied mathematics degree with pragmatic engineering that emphasizes observability, security, and measurable SLAs. An analytical problem-solver, he’s as likely to tune query planners and build load tests as he is to integrate ML model parsers for real-world inference pipelines.
13 years of coding experience
7 years of employment as a software developer
Bachelor’s Degree applied mathematics and computer science, Bachelor’s Degree applied mathematics and computer science at Moscow Power Engineering Institute (Technical University)Lomonosov Moscow State University
English
Stackoverflow
31reputation
3kreached
4answers
0questions
Github Skills (22)
machine-learning10
java10
apache-ignite10
javas10
refactoring10
data-structure9
algorithm9
algorithms9
performance-optimization9
vector9
data-structures9
data-analysis8
big-data8
software-design8
databases7
Programming languages (10)
Github contributions (5)
apache/ignite
Oct 2020 - Dec 2020
Apache Ignite
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:24 reviews, 5 commits, 5 PRs in 2 months
Contributions summary:Mark contributed to the Apache Ignite ML module by fixing a critical bug in the `Vector::kNorm` evaluation for odd powers, ensuring the correct calculation of vector norms. They added support for new distance metrics, including BrayCurtis, Canberra, JensenShannon, and WeightedMinkowski, enhancing the library's functionality for various data analysis tasks. Further contributions involved code cleanup in utility classes and the addition of a target encoding preprocessor, showcasing their involvement in improving the ML codebase. Additionally, the user integrated the Catboost model parser and examples.
data-management-platformin-memory-databasedistributed-sql-databaseapache-ignitenetwork-client
facebookarchive/nailgun
Aug 2020 - Aug 2020
Nailgun is a client, protocol, and server for running Java programs from the command line without incurring the JVM startup overhead.
Role in this project:Backend Developer
Backend DeveloperContributions:5 commits, 7 PRs in 1 day
Contributions summary:Mark primarily focused on refactoring and improving the codebase of the Nailgun server. They replaced `StringBuffer` with `StringBuilder` for performance optimization and removed a public constructor from `NGConstants`. Furthermore, they added an `Override` annotation to `NGExitException` for code clarity and implemented null checks for input arguments in the `NailStats` class. Lastly, they refactored `NGServerContext` to a static class.
java-programsoverheadjvmprogramsjar
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.