Svilen Kanev is a software engineer with 17 years' experience focused on computer architecture, systems software and cross-stack performance optimization, currently contributing at Google from San Francisco. His background blends deep academic research—PhD work and teaching at Harvard on power-efficient architecture and simulation—with hands-on engineering across datacenter power/performance and tooling. He has driven performance-focused contributions to prominent open-source projects such as pprof and DynamoRIO's drcachesim, improving profiling, ELF/kernel symbolization, and cache-simulation tooling. Comfortable refactoring build systems and heuristics as well as instrumenting low-level behavior, he brings a pragmatic researcher’s attention to measurable system improvements. Although happily settled at Google, he retains a broad curiosity spanning robotics, signal processing and compiler/runtime-level optimizations.
pprof is a tool for visualization and analysis of profiling data
Role in this project:
Back-end Developer & Performance Engineer
Contributions:16 reviews, 6 commits, 7 PRs in 4 years 9 months
Contributions summary:Svilen focused on optimizing the `pprof` tool's performance and refining its internal logic. They addressed issues related to function name pruning, specifically handling anonymous namespaces and operator overloading. The user also refactored heuristics within the ELF parsing and kernel symbolization, extending support for PIE kernels and handling kernel relocation symbols. Furthermore, they made improvements to the build process.
Contributions:20 commits, 13 PRs, 1 push in 1 year 11 months
Contributions summary:Svilen primarily contributed to the drcachesim project, a dynamic instrumentation tool platform, by implementing and enhancing features within the cache simulator. Their work involved adding tools for analyzing memory reuse time and instruction counts, demonstrating an understanding of performance profiling. They also refactored and isolated code, improved the integration of third-party tools, and addressed build system issues by adding configure flags, resulting in more robust and efficient analysis capabilities.
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.