Alexey Perevalov
Principal Software Engineer at Huawei
Moscow, Moscow, Russia
Summary
🤩
Rockstar🎓
Top SchoolAlexey Perevalov is a Principal Software Engineer with 11+ years building cross-platform, high-performance systems across Linux and Windows, currently enhancing Kubernetes for telecom requirements at Huawei. He brings deep kernel and networking experience from long-term roles at Samsung—where he led Tizen OS features, upstreamed kernel work, and contributed to netfilter—and earlier systems and backup software development at Acronis. A C/C++ specialist with solid experience in resource-constrained and distributed environments, Alexey has contributed NUMA-aware CPU and topology improvements to Kubernetes and scheduler-plugins, demonstrating practical expertise in container scheduling and system architecture. He is comfortable across the stack—kernel, system frameworks, and backend services—and routinely applies object-oriented design, UML, and SCRUM practices to ship robust, readable code. Based in Moscow, he pairs low-level performance tuning with pragmatic engineering leadership, often translating telecom-grade requirements into upstream open-source contributions.
11 years of coding experience
12 years of employment as a software developer
Engineer, Software engineering, Engineer, Software engineering at Московский Государственный Технологический Университет (Станкин)
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
scheduler10
system10
kubernetes10
go10
api10
architectures10
resource-management10
task-scheduler10
apidoc10
containerization10
architecture10
plugin10
job-scheduler10
kubernetes-pods10
testing8
Programming languages (7)
Github contributions (5)
kubernetes-sigs/scheduler-plugins
Nov 2020 - Aug 2021
Repository for out-of-tree scheduler plugins based on scheduler framework.
Role in this project:
Backend Developer
Backend DeveloperContributions:36 reviews, 11 commits, 6 PRs in 9 months
Contributions summary:Alexey primarily contributed to the development of out-of-tree scheduler plugins within the Kubernetes ecosystem. Their work involved modifying configuration files, generating code for new plugins, and integrating unit tests. Specifically, the user implemented a NUMA aware scheduler plugin, incorporating changes related to the NodeResourceTopologyMatchArgs, indicating a focus on resource management and scheduling optimization. Furthermore, their contributions included fixing example configurations to align with case sensitivity.
schedulerk8s-sig-schedulingout-of-tree
kubernetes/kubernetes
May 2020 - Aug 2021
Production-Grade Container Scheduling and Management
Role in this project:Back-end Developer & System Architect
Back-end Developer & System ArchitectContributions:47 reviews, 13 commits, 24 PRs in 1 year 3 months
Contributions summary:Alexey focused on enhancing the Kubernetes CPU management system. They refactored code to use `cadvisor.MachineInfo` for retrieving NUMA node information, replacing an older method. The user also modified the `podresources` API to incorporate topology and CPU ID information, indicating a deep understanding of resource allocation and system architecture within the Kubernetes ecosystem. These changes demonstrate a strong focus on performance optimization and the internal workings of Kubernetes resource management.
containersschedulingdockergradeproduction-grade
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.