Tasmia Rahman
Senior Engineer at Qualcomm
Austin, Texas Metropolitan Area United States
Summary
👤
Senior🎓
Top SchoolTasmia Rahman is a Senior Engineer based in the Austin area with four years of industry experience and a strong academic foundation from the University of Tennessee and BRAC University. At Qualcomm she progressed from Engineer to Senior Engineer working on performance-critical ML and compiler-level features, contributing to the TVM deep learning compiler stack with optimizations for Hexagon and support for quantized and mixed data types. She combines compiler and systems-level expertise (LLVM, RISC-V) with hands-on ML engineering, able to translate theoretical concepts into production-ready implementations under minimal supervision. Known for clear, productive communication and teaching experience, she brings mentorship and curriculum development skills from prior academic roles. Tasmia’s portfolio shows a practical focus on low-level optimization and embedded ML, reflected in her contributions to high-profile open-source projects that accelerate ML on specialized accelerators.
4 years of coding experience
Bachelor’s Degree, Computer Science, Bachelor’s Degree, Computer Science at Brac UniversityMaster's degree, Computer Science, Master's degree, Computer Science at University of Tennessee, Knoxville
Github Skills (12)
compiler10
hexagons10
machine-learning10
compiler-compiler10
python10
gpu9
deeplearning-ai9
deep-learning9
performance-analysis8
tensor8
performance-monitor8
f7
Programming languages (1)
Github contributions (4)
apache/tvm
Jun 2022 - Dec 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:10 reviews, 9 commits, 16 PRs in 5 months
Contributions summary:Tasmia primarily contributed to the development and optimization of machine learning operations within the TVM compiler stack, particularly focusing on the Hexagon architecture. They added and improved slice operations for the Hexagon platform, including add, subtract, and multiply operations, and implemented support for quantized elementwise operations. The user also integrated and tested resize2d and global average pooling 2D operations, along with other code changes, focusing on the support of different data types such as float16, uint8 and int8 and fixed chunk size layout.
metalvulkancompilertensoropencl
trahman-quic/tvm
Jun 2022 - Dec 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:94 pushes, 15 branches in 6 months
cpugpu-programminggpu-accelerationtvmdeep-learning
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.