About

Mayank Kowshik is a seasoned software leader with over 11 years of relevant experience in developing and shipping complex software libraries and applications for GPUs and Deep Learning accelerators. He is currently the Director of Deep Learning Inference Workflows and Applications at NVIDIA, where he leads the CUDA Developer Tools for Tegra SoCs. Mayank has hands-on experience in developing the Software Stack for Deep Learning accelerators and GPUs, Operating System Internals, Device Drivers, CPU and GPU debuggers, Parallel Programming, Computer Architecture, and Embedded Systems. He has also developed and validated CUDA Debugger software for Maxwell and Kepler GPUs on fmodel simulation, RTL emulation, and final silicon. In his previous roles, Mayank served as the Manager of the Operating Systems Interface team at NVIDIA and as the Principal Engineer/Manager of System SW Architecture at Intel's Artificial Intelligence Products Group. He has also been involved in the development and debugging of Linux and Windows kernel-mode device drivers for Quadro GPUs and QuadroPlex external GPU solutions. Mayank holds a Master of Business Administration (M.B.A.) degree from the University of California, Berkeley, Haas School of Business, an MS degree in Computer Engineering from the University of Texas at Austin, and a BE degree in Electronics and Communication from Delhi College of Engineering. He completed his high school education at Delhi Public School, where he studied Science and Computers. Mayank's technical proficiency includes Deep Learning, Mobile, Software Engineering, Python, Java, Android, AWS, Neural Networks, and MySQL. He is a talented and productive team leader who has hired and developed many talented teams throughout his career.