Abhik Sarkar

GPUs power almost every modern Python workload in machine learning, vision, and scientific computing. Yet most Python developers treat GPUs like “faster CPUs” and hope frameworks will handle performance automatically.

With the GPU boom and CUDA a high barrier to entry, Python has become the practical bridge between developer productivity and performance. This workshop teaches hardware-aware GPU programming for Python developers, with a focus on how performance is less about model code and more about data movement

Building your own minimalist NumPy like, ArryPy, it reveals the secret sauce behind high-performance array computing. In this hands-on workshop, you’ll start with a naïve Python loop, then progressively supercharge it: first pure-Python ufuncs, next Cython-typed memoryviews, and finally native C...

This talk explores the intersection of Python and Cython, emphasizing how understanding Cython can deepen one's knowledge of the Python Virtual Machine (PVM) and Python's performance characteristics.