Roman Penyaev is a Staff Engineer with 17 years of deep systems and storage expertise, currently building confidential computing and VM attestation tooling at IONOS. He specializes in low-level Linux kernel and virtualization work, contributing to high-profile open-source projects like io_uring (liburing) and fio and implementing io_uring support for Ceph. Roman has led cross-functional teams to deliver custom kernels, edge virtualization stacks, and hardware-software products—from QEMU/kernel patches at ZEDEDA to end-to-end STM32 firmware and PCB design at Skycharge. He blends hands-on kernel hacking with product-focused leadership, mentoring engineers and translating customer feedback into robust, deployable systems. A less obvious strength is his ability to operate at the software-hardware boundary, evidenced by kernel drivers, kdump-based debugging for edge devices, and a hobby project applying RL to game environments.
17 years of coding experience
18 years of employment as a software developer
Moscow Aviation Institute (National Research University)
Contributions:544 reviews, 42 commits, 192 PRs in 4 months
Contributions summary:Roman's contributions primarily revolve around enhancing the Edge Virtualization Engine (EVE) project's system stability and logging capabilities. They introduced a kernel dump collection mechanism using `kdump` and incorporated it into the build process, enabling detailed crash analysis. The user also improved the logging infrastructure by implementing ISO date formats for the kdump files and including dmesg output alongside the kernel dumps. Moreover, they optimized existing code and refactored ZFS parameters setting.
Contributions summary:Roman primarily contributed to the fio project by implementing and refactoring core utility functions related to pattern generation and verification, critical components for I/O testing. They introduced `strntol`, a string to long integer function, and a set of functions in `lib/pattern.c` to parse and format combined pattern inputs, adding flexibility to how test data is generated. Their work also involved refactoring existing pattern-related code, replacing older functions like `fill_pattern` with optimized versions and updating the verification process to utilize the new pattern parsing capabilities.
testingflexible
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.