CuPy & RAPIDS
Numba & Custom CUDA Kernels
Deep Learning Frameworks
Distributed & Scalable Computing

GPU-Accelerated AI

Custom CUDA Kernel for Convolutional Neural Nets

Writing custom kernels for CNN layers in Python.

By Michael Mallari

Wednesday, August 7, 2024

Art of the Possible With AI

People are going to use more and more AI. Acceleration is going to be the path forward for computing. These [are] fundamental trends.

Jensen Huang, Nvidia Co-Founder & CEO, Circa 2025

NVIDIA CUDA Python Expert

Michael Mallari, MS, MBA

Michael is an adjunct faculty member at Columbia University—teaching cloud computing, data & analytics, and AI. A computer scientist-by-trade and an MBA, he brings a combined 20+ years of industry hands-on experience in software engineering, product design, product management, and data science—having created and captured business value with AI, 5G, messaging, fintech, martech, mobile, and even Web 1.0 solutions (at the start of his career during the dot-com boom).

Thriving at the intersection of emerging technologies, business strategy, and product sense, Michael has been entrusted with discovering, defining, delivering, and deploying next-gen solutions for Fortune 500 companies, startups, and not-for-profit organizations, including the following:

Michael has been a long-time ACM and SIGGRAPH member—dating back to his undergraduate studies, when he earned his BS in Computer Science from New York Institute of Technology. Subsequently, he earned his MBA from the University of Maryland, College Park, and MS in Applied Analytics from Columbia University. He is NVIDIA certified in GPU-accelerated AI.

Subscribe to Substack newsletter

Designed and built by Michael Mallari

Discover first-mover advantages based on newly published AI research. Read case studies »