Saman Ashkiani
Senior Software Engineer at NVIDIA
Redwood City, California, United States
Summary
👤
Senior🎓
Top SchoolSaman Ashkiani is a Senior Software Engineer with a decade of experience specializing in high-performance GPGPU computing, currently working at NVIDIA after roles at Google and OmniSci. He holds a PhD focused on primitive algorithms and dynamic data structures tuned for modern GPU architectures, and brings deep expertise in CUDA, parallel algorithms, and HPC systems. His work spans compiler techniques (JIT/LLVM), OLAP database internals, and performance optimization for ML frameworks, with notable open-source contributions to HeavyDB (OmniSci) that touch GPU memory handling and query engine features. Based in Redwood City, he combines rigorous academic training with production engineering, and is particularly energized by low-level performance problems and architecting concurrent, scalable systems.
10 years of coding experience
9 years of employment as a software developer
Master of Science - MS, Electrical and Computer Engineering, Master of Science - MS, Electrical and Computer Engineering at University of California, DavisBachelor of Science (B.Sc.), Electrical Engineering, Bachelor of Science (B.Sc.), Electrical Engineering at Sharif University of Technology
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (14)
databases10
c-language10
cprogramming-language10
relational-databases10
database-design10
sql10
sql-database10
database10
performance-optimization9
cuda9
gpu-programming9
data-structures8
data-structure8
thrift8
Programming languages (3)
Github contributions (5)
heavyai/heavydb
May 2018 - Apr 2020
HeavyDB (formerly OmniSciDB)
Role in this project:
Back-end Developer & Database Engineer
Back-end Developer & Database EngineerContributions:97 commits, 3 comments in 1 year 11 months
Contributions summary:Saman contributed to the database layer of the HeavyDB repository, adding features for partition details in the `\d` command, introducing new enum members to the Thrift definitions (`TPartitionDetail`) and updating the table details structure (`TTableDetails`). The user also implemented code modifications to the C++ back-end of the database in the `SQLFrontend`, `ThriftHandler`, and `QueryEngine` directories. The user worked on features such as inner-join-fragment-skipping and GPU related memory handling and sync for Volta architecture.
cudareal-timevisualizationgpuheavyai
owensgroup/SlabHash
Jan 2019 - Jul 2019
A warp-oriented dynamic hash table for GPUs
Contributions:77 commits, 4 PRs, 5 pushes in 5 months
cudampihashgpuwarp
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.