Implement and optimize Machine learning training & inference and Computer Vision

related algorithms for the target GPU / DSP / AI Accelerator (NPU) hardware

? Vectorize various algorithms using hardware-specific ISA (intrinsic/assembly) and achieve

close to theoretical peak performance

? Come up with algorithm designs in such a way that it leverages the full potential of the

target hardware architecture

? Implement testbench for the hand-optimized algorithms to test quality and performance

Required skills

? BTech/BE/MTech/ME/MS/PhD degree in CSE/IT/ECE

? > 2 years of experience working in Algorithm Development, Porting, Optimization &

Testing

? Strong Knowledge of Computer Architecture

? Excellent proficiency in C and C++ programming

? Proficiency in Assembly / Intrinsic (SIMD/VLIW)

? Good knowledge of DMA, Memory banks, and cache concepts

? Good understanding of machine learning / Deep learning and exposure to DL model

architectures

? Strong problem-solving, analytical, and debugging skills