Ariel Ben-yehuda
Head Of Research at Stealth
Israel
Summary
🤩
Rockstar🎓
Top SchoolAriel Ben-yehuda is a research-oriented software leader with 12 years of experience building high-performance systems and compilers, currently serving as Head of Research at a stealth company in Israel. He previously helped ship ElastiCache Serverless at AWS and spent formative years on the Rust compiler team at Mozilla, where his work reduced stack usage and advanced the type/borrow-checker. Ariel is an active performance engineer and backend developer in open source—contributing to async-profiler (adding Rust symbol demangling and precise TSC timing) and improving rustc-perf’s benchmarking infrastructure. He blends low-level systems expertise with production-grade cloud experience, often tackling profiling, tooling, and performance regressions that are hard to reproduce. Colleagues know him as a pragmatic mentor who integrates diverse contributions without introducing regressions. He holds a BSc in Mathematics and Computer Science from Tel Aviv University.
12 years of coding experience
11 years of employment as a software developer
BSc Mathematics and Computer Science, BSc Mathematics and Computer Science at Tel Aviv University
Github Skills (16)
performance-monitor10
rust10
c-language10
performance-monitoring10
performance-analytics10
cprogramming-language10
performance-analysis10
code-profiling10
profiling10
java9
jfr9
makefile9
gnu-make9
javas9
version-control7
Programming languages (10)
Github contributions (5)
async-profiler/async-profiler
Dec 2024 - Apr 2025
Sampling CPU and HEAP profiler for Java featuring AsyncGetCallTrace + perf_events
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:21 reviews, 5 PRs, 42 comments in 4 months
Contributions summary:Ariel primarily contributed to the core functionality and performance optimization of the async-profiler. Their work included implementing demangling of Rust symbols and integrating the Time Stamp Counter (TSC) for accurate time measurements. They also added a new API `asprof_get_thread_local_data` to collect per-thread profiling data and fixed an issue related to setting the clock in the JFR recording.
java
rust-lang/rustc-perf
Apr 2017 - Aug 2017
Website for graphing performance of rustc
Role in this project:Backend Developer
Backend DeveloperContributions:15 commits in 3 months
Contributions summary:Ariel refactored the benchmark code, enabling separate benchmarking of commits for debugging purposes. This involved changes to the `benchmark.rs` file, suggesting modifications to the core performance testing logic. The user also split scripts into modules and fixed various issues within the codebase. These changes indicate a focus on improving the maintainability and functionality of the performance collection and analysis tools.
rustperformancegraphrustcgraphing
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.