Javed Absar
Principal Engineer, ML AI Compilers Research & Engineering at Qualcomm
Cambridge, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolJaved Absar is a Principal Engineer specializing in ML/AI compilers and compiler research with 11+ years of hands-on experience across industry leaders including Qualcomm, Graphcore, ARM and AMD. He holds a PhD from KU Leuven/IMEC in compilers and architectures and has driven production MLIR and Triton compiler efforts, while also pioneering Graphcore’s move to MLIR. Author of 15 US patents and 15+ papers, co-author of a Springer book on ultra low-power DSPs, he combines deep academic rigour with practical compiler engineering for CPUs, GPUs and IPUs. Notably, his open-source contributions extend Clang’s ARM support for Cortex-R series and half-precision FPUs, reflecting a knack for low-level target work that materially improves toolchains. Based in Cambridge, UK, he blends research leadership, technical architecture and product-focused delivery across both silicon and software stacks.
11 years of coding experience
23 years of employment as a software developer
PhD Compiler and Architecture, PhD Compiler and Architecture at KU LeuvenMBA Management, MBA Management at Vlerick Business SchoolComputer Engineering, Computer Engineering at Nanyang Technological University Singapore
Github Skills (14)
c1710
compiler10
arm10
compiler-compiler10
cpp10
architectures10
c1110
cplus10
architecture10
sys9
embedded9
performance-analysis8
performance-monitor8
test-automation7
Programming languages (2)
Github contributions (5)
llvm-mirror/clang
Apr 2015 - May 2019
Mirror kept for legacy. Moved to https://github.com/llvm/llvm-project
Role in this project:
Embedded Systems Engineer / Automation Engineer
Embedded Systems Engineer / Automation EngineerContributions:11 commits in 4 years 2 months
Contributions summary:Javed's primary contributions focused on extending Clang's support for ARM architecture, specifically targeting Cortex-R4/R4F and later Cortex-R52 processors. They implemented and tested support for new ARM -mfpu options related to half-precision floating-point and vfpv3xd. Additionally, they were involved in fixing issues related to vector alignment and test cases within the Clang compiler infrastructure. These contributions involve modifying the driver and preprocessor for ARM targets, and also enhancing target support.
keptwindowsllvmcc-plus-plus
mono/llvm
Apr 2015 - May 2019
Mono LLVM Repository.
Contributions:123 commits in 4 years 1 month
database-accessmonocompilerreductionllvm
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.