Abolfazl Shahbazi
AI Frameworks Engineer at Intel Corporation
Portland, Oregon, United States
Summary
🤩
Rockstar🎓
Top SchoolAbolfazl Shahbazi is an AI Frameworks Engineer with 10+ years of experience optimizing and packaging deep learning workloads for cloud and bare-metal environments, currently building high-performance AI tooling at Intel in Portland. He combines hands-on performance engineering—benchmarks, profilers, and Intel-optimized TensorFlow/PyTorch distributions—with Kubernetes and Kubeflow expertise to move models from research to production. His background spans big data and cloud platform engineering, where he led Spark/YARN tuning, Hadoop automation, and CI/CD improvements, giving him a rare full-stack view of data-to-inference pipelines. An active open-source contributor, he has improved tensorflow/serving and kubeflow build and deploy processes and helped harden Intel AI reference models for production use. He also brings security-minded engineering practices and mentorship experience, and has a practical knack for turning complex performance issues into repeatable, distributable artifacts like Docker images and Python wheels.
10 years of coding experience
13 years of employment as a software developer
Amirkabir University of TechnologyComputer Science, Computer Science at Portland State UniversityMathematics, Mathematics at Sharif University of Technology
Persian
Stackoverflow
171reputation
42kreached
5answers
0questions
Github Skills (29)
kubernetes10
python10
machine-learning10
inference10
gcp10
deeplearning-ai10
deep-learning10
tensorflow10
deep-neural-networks10
jsonnet10
kubernetes-pods10
serve10
bash9
devops9
protobuffer9
Programming languages (18)
Github contributions (5)
intel/ai-reference-models
Nov 2018 - Dec 2022
Intel® AI Reference Models: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors and Intel® Data Center GPUs
Role in this project:
ML Engineer
ML EngineerContributions:17 releases, 5 reviews, 813 commits in 4 years 1 month
Contributions summary:Abolfazl primarily focused on improving the performance and functionality of the Intel AI Reference Models, which involve running deep learning workloads. Their contributions included adding Intel Apache license headers to the code, moving encoding definitions for better PEP263 compatibility, and fixing bash scripting issues. In addition, they integrated a fix for Pillow version for the unet model and added libsnd dependencies for the wavenet model, ensuring the compatibility and stability of the model.
optimizationsprocessorstensorflowzoomodel-zoo
kubeflow/kubeflow
Sep 2018 - Feb 2019
Machine Learning Toolkit for Kubernetes
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:13 commits, 22 PRs, 133 comments in 5 months
Contributions summary:Abolfazl primarily contributed to the Kubeflow project by addressing code style issues, fixing parameters, and enhancing code in JSONnet files related to MXNet and PyTorch jobs. The user also worked on improving the build process by adding a `--skipInitProject` option and correcting errors in `kfctl.sh` related to GCP project initialization. Furthermore, the user was involved in updating cluster versions and replacing cloud parameters with more direct names, demonstrating skills in infrastructure configuration and build process management.
pythondata-sciencenotebookmachine-learningmlops
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.