Tal Ben-nun
Scientific Advisor at Lawrence Livermore National Laboratory
United States
Summary
👤
Senior🎓
Top SchoolTal Ben-nun is a research staff scientist with 17 years of experience specializing in scalable machine learning, high-performance computing, and programming languages, currently working at Lawrence Livermore National Laboratory and advising at Daisytuner. His work bridges large-scale distributed ML for scientific and weather/climate applications, data-centric parallel programming (notably contributions to the DaCe project), and machine comprehension of code through deep learning and partial compilation. He has a strong systems background from low-level GPU and multi-GPU programming to Linux kernel networking and nonlinear optimization, informed by postdoctoral research at ETH Zurich and a PhD from Hebrew University. Tal routinely supervises graduate research that leads to publications and awards, and his practical contributions to parsing and analysis in DaCe reflect a rare mix of compiler-level insight and applied ML for scientific workflows.
17 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at The Hebrew University of Jerusalem
English, Hebrew
Github Skills (15)
wiiu10
g10
wii10
c-language10
cprogramming-language10
parsing10
emulation10
python10
parse10
cube10
manipulation10
code-generation9
file-handling9
api8
apidoc8
Programming languages (15)
Github contributions (5)
spcl/dace
Feb 2019 - Dec 2022
DaCe - Data Centric Parallel Programming
Role in this project:
Back-end Developer
Back-end DeveloperContributions:23 releases, 1183 reviews, 4329 commits in 3 years 10 months
Contributions summary:Tal primarily contributed to the DaCe project by modifying code within the Python frontend, with a focus on improving the parsing and analysis of Python code. Their contributions included updates to the context manager inlining, preprocessing steps, and enhancements to handle string literals and complex structures in callbacks. The user's work also involved adjustments to symbol handling, specifically with respect to nested SDFGs and constant parameters.
cudahigh-level-synthesisparallelhigh-performance-computingvivado-hls
tbennun/nccl-windows
Jun 2016 - Jan 2017
Contributions:3 commits, 2 PRs, 2 pushes in 7 months
visual-studiowindowsgpucommunicationoptimized
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.