Summary
Kashyap Garimella is a Senior GPU Architect with 15 years of deep experience across compiler design, CPU/GPU/accelerator architecture, and systems-level engineering. He has led code generation, simulation, and performance work for AI processors at Habana Labs and now develops proto compilers and kernels for future NVIDIA architectures. His background includes low-level CPU decoder and binary translation performance optimization at NVIDIA, plus MLIR-based fusion and lowering of PyTorch graphs, reflecting a rare blend of microarchitecture insight and modern compiler toolchain expertise. Based in Bengaluru and trained at IIT Madras, he pairs rigorous academic foundations with hands-on post-silicon validation and target-aware fuzzing experience. Notably, he moves fluidly between writing simulators and shaping high-level MLIR transforms, making him effective at bridging hardware realities and compiler abstractions.
15 years of coding experience
12 years of employment as a software developer
Indian Institute of Technology Madras