Ashkan Aliabadi

San Francisco, California, United States

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Ashkan Aliabadi is a software engineer with six years focused on high-performance computing, on-device AI, computer graphics, and embedded systems, based in San Francisco. He led CPU and GPU backend and performance work for PyTorch Mobile at Facebook AI Infra and previously founded a cross-platform, inference-only DNN library to avoid vendor lock-in for embedded devices. His background spans firmware for Intel SoCs, AR-focused computer vision at Magic Leap, and real-time rendering engines, giving him a pragmatic blend of low-level optimization and applied ML. An active open-source contributor, he’s improved ARM detection and cross-platform init routines in the widely used pytorch/cpuinfo project, reflecting deep expertise in ARM architectures and portable performance. Notably, he combines assembly- and Vulkan-level optimizations with product-minded engineering to deliver fast, deployable on-device AI.

6 years of coding experience

15 years of employment as a software developer

Bachelor of Engineering - BE, Computer Software Engineering, Bachelor of Engineering - BE, Computer Software Engineering at Shiraz University

Advanced 3D Computer Graphics, Advanced 3D Computer Graphics at Linköping University

Github Skills (10)

c1710

arm10

architectures10

c1110

architecture10

linux9

windows8

emscripten7

broadcom7

cpu5

Programming languages (4)

C++CObjective-CPython

Github contributions (5)

pytorch/cpuinfo

Feb 2020 - Jul 2021

CPU INFOrmation library (x86/x86-64/ARM/ARM64, Linux/Windows/Android/macOS/iOS)

Role in this project:

Back-end Developer

Contributions:7 commits, 16 PRs, 7 pushes in 1 year 5 months

Contributions summary:Ashkan's contributions primarily involve updating and improving the `cpuinfo` library, focusing on ARM architecture-related code. Their work includes adding support for new Broadcom and Texas Instruments chipsets, specifically implementing functions to decode and identify them. They also made changes to the Emscripten and Windows initialization routines, indicating a focus on cross-platform compatibility and integrating the library with different build environments. Several commits focused on adapting the library to handle different Raspberry Pi hardware revisions and improve existing ARM CPU detection and cache information.

kernelcpu-topologyiosarm64cpu

AshkanAliabadi/pytorch

Aug 2019 - Jul 2020

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Contributions:370 pushes, 40 branches, 2 comments in 11 months

pythongpu-accelerationdeep-learninggpuacceleration

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.

Request Free Trial