Alessandro Morari

Alessandro Morari is an AI systems leader at NVIDIA in the DevTech AI organization. His current focus is on AI-driven GPU kernels and next-generation programming models for accelerated computing. His experience spans the full AI stack, from GPU kernel optimization to AI product leadership. Before NVIDIA, he led the team at IBM Research that shipped the Watson Code Assistant, one of the earliest large-scale generative AI products. He previously worked on system software for the Summit and Sierra supercomputers and created NYU Courant's first course on high-performance machine learning. Morari has authored over 30 publications, holds 15 patents, and earned a Ph.D. in Computer Architecture.
Avatar photo

Posts by Alessandro Morari

Decorative image.
Developer Tools & Techniques

Tuning Flash Attention for Peak Performance in NVIDIA CUDA Tile

In this post, we dive into one of the most critical workloads in modern AI: Flash Attention, where you’ll learn: How to implement Flash Attention using NVIDIA... 20 MIN READ