Iman Tabrizian
Senior Deep Learning Software Engineer at NVIDIA
United States
Summary
🤩
Rockstar🎓
Top SchoolIman Tabrizian is a Senior Deep Learning Software Engineer with 11 years of experience, currently advancing production inference at NVIDIA as a core contributor to the Triton Inference Server. With an MASc from the University of Toronto and a strong background in distributed ML and network automation, she blends research rigor from roles at Vector Institute and UofT with hands-on systems engineering in C++, Python, and PyTorch. Her open-source work on Triton’s Python backend, client libraries, and testing/CI improvements demonstrates a focus on robust, high-performance inference pipelines and GPU/shared-memory integrations. She’s as comfortable optimizing low-level C-API and CUDA paths as she is adding reliability to test automation—an uncommon mix that accelerates ML from prototype to scalable deployment.
11 years of coding experience
2 years of employment as a software developer
Amirkabir University of TechnologyMaster of Applied Science (MASc), 3.94/4.0, Master of Applied Science (MASc), 3.94/4.0 at University of Toronto
Stackoverflow
11reputation
653reached
0answers
1question
Github Skills (38)
unit-testing10
client-library10
performance-monitor10
c-language10
performance-analytics10
python10
code-profiling10
scripting10
testing10
numpy10
triton10
script10
performance-analysis10
profiling10
sh10
Programming languages (20)
Github contributions (5)
triton-inference-server/python_backend
Aug 2020 - Jan 2023
Triton backend that enables pre-process, post-processing and other logic to be implemented in Python.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:568 reviews, 123 commits, 247 PRs in 2 years 4 months
Contributions summary:Iman's commits primarily involve the development and maintenance of the Python backend for Triton Inference Server. They are responsible for the initial setup and subsequent improvements to the Python backend, implementing core functionalities such as handling requests, managing input/output tensors, and integrating with the underlying Triton server. The user's work spans both the core backend logic and enhancements, including features for GPU tensor support and interaction with the shared memory. This user is focused on building the core inference pipeline.
post-processingpre-processpythonbackendpython3
triton-inference-server/client
Jul 2020 - Jan 2023
Triton Python, C++ and Java client libraries, and GRPC-generated client examples for go, java and scala.
Role in this project:Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:303 reviews, 37 commits, 81 PRs in 2 years 5 months
Contributions summary:Iman primarily contributed to the client libraries, specifically the Python client, for the Triton Inference Server. Their work involved fixing bugs related to data serialization, particularly for the BYTES data type, and addressing issues with trailing zeros. Additionally, the user introduced and improved performance measurement features, including count-based stabilization, and made modifications to the C-API to support CUDA and system memory, which likely involved performance profiling and optimization.
golangpythongrpcscalaclient-libraries
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.