Andrei Gudkov is a Principal Researcher based in Moscow with 9 years of focused experience building high-performance, algorithmically complex server software for cloud, AI training and big data systems. He designs and implements production schedulers and consolidation algorithms for Huawei Cloud that deliver measurable gains (up to 18.7% host reduction and 4.5% LLM training speedups) using NUMA-awareness, stochastic optimization and migration-duration prediction. Previously he led backend architecture and search teams at startups and major web companies, shipping large-scale vertical search, distributed datastores and latency-critical pipelines. A hands-on performance engineer, he contributes to the widely used CRoaring roaring-bitmap project in C with SIMD optimizations and has implemented range/removal primitives and critical efficiency fixes. Comfortable spanning research, core systems design and low-level optimization, he pairs theoretical bounds and simulations with pragmatic production fixes. Colleagues rely on him for solving knotty resource-allocation and storage problems where both correctness and throughput matter.
Roaring bitmaps in C (and C++), with SIMD (AVX2, AVX-512 and NEON) optimizations: used by Apache Doris, ClickHouse, Alibaba Tair, Redpanda, YDB and StarRocks
Role in this project:
Back-end Developer & Performance Engineer
Contributions:18 commits, 12 PRs, 1 push in 1 year 3 months
Contributions summary:Andrei primarily focused on enhancing the performance and functionality of the roaring bitmap library. They introduced a new function `roaring_read_uint32_iterator()` and benchmarked its performance, alongside the existing add functions. Furthermore, the user implemented `roaring_bitmap_remove_range()` and `roaring_bitmap_remove_many()`, extending the library's capabilities. They also optimized several container equality functions to improve efficiency and fixed critical bugs related to range operations and memory management, contributing to the stability and reliability of the library.
Contributions:1 release, 7 pushes, 1 branch in 2 years 4 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.