Tareq Malas - Research Scientist ML Systems at Meta

Tareq Malas

Research Scientist ML Systems at Meta

Thuwal, Makkah Region, United States

Join Prog.AI to see contacts

Summary

👤

Senior

🎓

Top School

Tareq Malas is a Research Scientist specializing in ML systems and high-performance computing with 11 years of experience optimizing large-scale scientific and recommendation workloads. At Meta he drives hardware/software co-design and cross-stack quantization to accelerate recommendation models on Meta’s MTIA accelerators, and he builds benchmarking and profiling frameworks that inform system-wide optimizations. His background includes enabling NAMD and Amber for exascale and production supercomputers at Intel and applied research on manycore processors at Berkeley Lab, giving him deep expertise across CPUs, GPUs, and custom accelerators. He combines low-level performance modeling and pre-silicon studies with practical runtime and compiler collaboration to remove system bottlenecks. Passionate about applying efficiency gains to science, climate, and healthcare, he brings a rare blend of HPC domain knowledge and production ML systems engineering. An often-overlooked strength is his track record of translating research-grade performance insights into deployable tooling that scales across massive infrastructures.

11 years of coding experience

9 years of employment as a software developer

BS, Computer Engineering, BS, Computer Engineering at AL-Balqa’ Applied University

PhD, Computer Science, High Performance Computing, PhD, Computer Science, High Performance Computing at King Abdullah University of Science and Technology

BS, Computer Engineering, BS, Computer Engineering at University of Jordan

Arabic, English

Github Skills (15)

multiplication9

matrix8

harness8

facebook8

optimization7

matrix-multiplication7

hardware4

stencil3

pde-solver3

pde2

neural-network2

petsc2

python2

compiler2

partial-differential-equations1

Programming languages (3)

C++CFortran

Github contributions (5)

ecrc/girih

Sep 2014 - Aug 2016

This tool serves as a test harness for different optimization techniques to improve stencil computations performance in shared and distributed memory systems

Contributions:8 commits, 256 pushes, 1 branch in 1 year 11 months

distributeddistributed-memorymemorytechniquestest-harness

azrael417/caffe

Oct 2016 - Dec 2016

Contributions:41 pushes, 5 branches in 2 months

deep-learningpytorchcaffemachine-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.

Request Free Trial