Aadesh Deshmukh
Sr. Deep Learning Compiler Engineer at AMD
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolAadesh Deshmukh is a Sr. Deep Learning Compiler Engineer with 8 years of experience building high-performance ML systems, GPU kernels, and compiler toolchains for production AI at AMD and national labs. He specializes in HW/SW co-design and memory-hierarchy aware compilers for GEMM, convolution, and attention workloads, delivering automated op-shape mapping, layer fusion, and software pipelining for VLIW-SIMD cores. His background spans scientific computing and HPC—optimizing torch/tvm pipelines for reactor simulations, launching thousands of MFIX simulations, and accelerating finite-volume solvers with 10–20x GPU speedups. Aadesh combines research rigor (M.S. CS, 3.86 GPA) with practical production validation across architecture, emulation, and quantization teams, and maintains an engineering blog while balancing life with biking, standup comedy, and hiking. An uncommon strength is translating domain-specific DSL and MLIR-based ideas into deployed compiler optimizations that materially reduce memory roundtrips for inference.
8 years of coding experience
4 years of employment as a software developer
The University of UtahB.TECH., Computer Science and Engineering, B.TECH., Computer Science and Engineering at Indian Institute of Information Technology, Tiruchirappalli
English, Hindi, Marathi, Telugu
Github Skills (61)
anomaly-detection9
polygons9
data-mining9
convex-hull9
calculus9
forecasting9
approximation8
julia8
factorization8
polyhedra8
scikit-learn8
machine-learning8
time-series8
crossword8
ssreflect8
Programming languages (7)
Github contributions (5)
alan-turing-institute/TimeSeriesClassification.jl
May 2020 - Dec 2020
Machine Learning with Time Series in Julia
Contributions:80 commits, 11 PRs, 80 pushes in 6 months
statisticsmachine-learningjuliatime-seriestimeseries
aa25desh/TaylorModels.jl
May 2019 - Oct 2020
Rigorous function approximation using Taylor models in Julia
Contributions:1 PR, 23 pushes, 3 branches in 1 year 4 months
function-approximationapproximationrigorousjulia
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.