Summary
Evgueni Brevnov is a Senior Software Engineer with over 20 years of deep systems and compiler experience, currently focused on JVM and LLVM-based compiler optimizations at Azul Systems. He has repeatedly delivered measurable performance wins—up to 30% on micro-benchmarks and large customer app speedups of 25%—by implementing low-level x86 optimizations, calling-convention improvements, and novel intrinsic handling. Previously at Intel he led vectorizer and SIMD modeling work, designed loop-peeling and reassociation passes, and removed compile-time bottlenecks that yielded multi-hundred-percent improvements in real workloads. Evgueni combines strong C++/Java expertise with a rigorous understanding of memory models and multithreading, having fixed subtle volatile-memory races that prevented data loss in customer systems. Based in Novosibirsk and trained in applied mathematics and computer science at NSU, he brings a rare blend of compiler research, pragmatic engineering, and production-grade performance tuning. A pattern in his career is turning difficult low-level correctness and performance problems into deployable compiler features used by external customers.
6 years of coding experience
17 years of employment as a software developer
Master of Science (MS), Applied Mathematics and Computer Science, Master of Science (MS), Applied Mathematics and Computer Science at Novosibirsk State University (NSU)
English, Russian