Summary
Ankur Deshwal is a Staff Engineer with over 15 years in the semiconductor industry, specializing in GPGPU, mobile-AI accelerators, and optimizing compilers for high-performance and deep-learning workloads. Based in Bengaluru, he leads R&D efforts to overcome the memory wall, building drivers, LLVM backends, and cycle-accurate/approximate simulators for tile-based mobile GPUs and custom caches. His work spans end-to-end stack: OpenCL drivers, GLSL frontends, compiler toolchains and architectural verification, with prior roles at SAIT, Samsung Electronics and an internship at NVIDIA. A Machine Learning Nanodegree and deep-learning/GANs specializations complement his hands-on hardware/software co-design focus, and he’s known for pragmatic prototypes that turn research ideas—like experimental ray-tracing compilers and scalable mobile GPU drivers—into production-ready tooling.
11 years of coding experience
5 years of employment as a software developer
Bachelor of Technology (B.Tech.), Electronics and Communications Engineering, Bachelor of Technology (B.Tech.), Electronics and Communications Engineering at JSS Academy of Technical Education
Machine Learning Nanodegree, Machine learning, Machine Learning Nanodegree, Machine learning at Udacity
Online Specialization - Deep Learning, Online Specialization - Deep Learning at Coursera, DeepLearning.ai
Indian Institute of Technology Kanpur
Online specialization - MLOps, Online specialization - MLOps at Coursera, deeplearning.ai