Pradeep Ramani
Senior Deep Learning Architect at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolPradeep Ramani is a Senior Deep Learning Architect based in California with nine years of focused experience accelerating GPGPU compute for machine learning and graphics. He has driven foundational work at NVIDIA (contributing to CUTLASS and libraries used by cuDNN/cuBLAS/TensorRT) and previously shaped Intel GPU microarchitecture and feature sets across multiple generations via simulators, micro-benchmarks and hardware/software co-design. His expertise spans CUDA, OpenCL, signal processing, and hardware verification, enabling order-of-magnitude performance and power improvements through cross-stack optimization. Equally comfortable in low-level RTL/testbench development and high-performance linear algebra kernels, he blends academic rigor (MS, UC Santa Barbara, 4.0 GPA) with production shipping at scale. A lesser-known strength is his track record of building the very performance test suites and micro-benchmarks that informed silicon decisions, giving him rare insight into both silicon-level tradeoffs and ML library performance.
9 years of coding experience
7 years of employment as a software developer
Anna University, ChennaiM.S., Computer Engineering, Signal and Image Processing., CGPA : 4.0, M.S., Computer Engineering, Signal and Image Processing., CGPA : 4.0 at University of California, Santa Barbara
Github Skills (11)
cuda10
matrix-multiplication10
nvidia10
cluster-computing10
deep-learning10
cpp10
blas10
gpu10
linear-algebra10
algebra10
tensor9
Programming languages (1)
Github contributions (5)
IonThruster/vcdgen
May 2020 - Aug 2021
Contributions:14 commits, 11 pushes, 1 branch in 1 year 3 months
IonThruster/CudaTutorials
Oct 2019 - Mar 2024
Contributions:4 pushes, 1 branch in 4 years 5 months
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.