Welcome to Harisankar Sadasivan’s Homepage

Senior AI Performance Engineer, NVIDIA | Affiliate Assistant Professor, UW Seattle | Distinguished Visitor, IEEE-CS

Welcome to Harisankar Sadasivan’s Homepage

Senior AI Performance Engineer, NVIDIA | Affiliate Assistant Professor, UW Seattle | Disintiguished Visitor, IEEE-CS

Hi, I'm Hari Sadasivan

Senior AI Performance Engineer, NVIDIA | Affiliate Assistant Professor, UW Seattle | Distinguished Visitor, IEEE-CS

Sarah Chen
Seattle, WA, USA

Dr. Harisankar (Hari) Sadasivan is a Senior AI Performance Engineer at NVIDIA, an affiliate assistant professor at the University of Washington in Seattle, and a Senior Member and Distinguished Visitor of the IEEE Computer Society. Hari’s research centers on three Critical and Emerging Technologies designated as vital to U.S. national security: advanced computing, artificial intelligence, and biotechnologies. Hari currently leads NVIDIA’s cuEquivariance library's GPU-accelerated AI kernels for drug discovery. His collaborators include a Nobel Laureate, leading researchers at institutions such as MIT and Stanford, and several frontier AI startups. Previously, Hari led the AMD Center of Excellence in AI at the University of Washington, contributed to AMD Composable Kernel library (used by tech-giants including Meta and MS to accelerate AI), and co-founded AMD’s life sciences group, where he helped build high-performance AI solutions at the intersection of computing and biology.

Hari received his PhD and Master’s degrees in Computer Science and Engineering from the University of Michigan, Ann Arbor, where he was advised by Prof. Satish Narayanasamy. His doctoral research advanced the state of the art in hardware–software co-design for accelerated and portable long-read DNA sequencing. Hari’s PhD work received the prestigious 2022 MICRO Top Picks Honorable Mention, along with multiple artifact badges recognizing its rigor, impact, and reproducibility. He serves as a technical reviewer for several leading IEEE and ACM venues, including some of the longest-running and most influential conferences and journals in computer architecture and systems. Hari has previously brought his expertise to industry roles at Samsung R&D.

"

"Nothing in life is to be feared, it is only to be understood"- Marie Curie

Teaching

Industry meets academia to define and close the gaps between AI HW, SW and students

Teaching

I joined the University of Washington's faculty in December 2023 and am currently an Affiliate Assistant Professor in ECE.AI’s demand for FLOPs has far outpaced what Moore’s Law could offer. This gap can only be bridged with aggressive software innovations. In order to do this, it is necessary to understand the AI software stack and how it maps to the GPU hardware. I designed and teach a course on AMD GPU Programming at the University of Washington Seattle. The course projects informed several HPC and AI projects at AMD.

To further the initiative of understanding AI, I digged into today's workhorses of AI- the GPUs. This led me to design a new course titled “Matrix to Machines: GPU HW Design on FPGA for AI”, where students designed a GPU hardware and software stack from scratch on an FPGA.

Advising

Past students whom I co-advised include:
-Melissa Queen, University of Washington Seattle
-M. Emin Ozturk, University of Utah
-Juechu Dong and Xueshen Liu , U Michigan Ann Arbor

News

Latest updates.

[December 2025] — Symposium Chair
I'm chairing an IEEE joint-symposium on Systems for AI and robotics in Seattle.

[December 2024] — Invited to speak at the US Dept. of Defense
I was invited to address the challenges in HW-SW for the future of AI and genomics by the Defense Intelligence Agency, Department of Defense, USA.

[Nov 2024] — Minimap2 Chaining – ACM BCB 2024
Our GPU-accelerated Minimap2 chaining paper was presented at ACM BCB 2024. Here’s an AMD blog post on the collaboration.

[August 2024] — Speeding up AI matmuls on GPUs
Our work on improving work-partitioning for AI on GPUs is out. Here’s a pre-print.

[July 2024] — Defining the future of genomics HW and SW
Our paper on setting the direction of genomics acceleration for the coming decades is out. Here’s a pre-print.

Reseach

Research

I envision a world where AI is advanced and performant enough to diagnose and find cures for all human diseases via techniques such as Precision Medicine & Drug Discovery. I realize that’s a big jump. So, I have broken down my interests for now.

High Performance Artificial Intelligence (AI):

Tall & Skinny GEMMs, Stream-K, Performance issues on multi-chiplet GPUs, LLM inference optimizations, Faster attention kernels

Research

High Performance -omics & Drug Discovery:

Long-read DNA alignment, Raw signal alignment, AI-based Basecalling, Metagenomics, Protein Structure Prediction for drug discovery

Recent Publications

Research contributions across genomics, GPU acceleration, computational biology, and high-performance computing.

  • All
  • Pre-print
  • Journal
  • Conference
  • Workshop
Genomic Computing Revolution
Pre-print 2024

Adaptive GPU GEMM Kernel Scheduling and Selection using Bloom Filters

Harisankar Sadasivan, Muhammad Osama, Maksim Podkorytov et al. In submission.

Genomic Computing Revolution
Journal 2024

brain lymphoma diagnostics through nanopore sequencing of cytology-negative CSF

J. Hench, C. Hultschig, I. Bratic Hench, Harisankar Sadasivan et al. Acta Neuropathologica. Springer.

Genomic Computing Revolution
Conference 2024

Genomic Computing Revolution: Defining the Next Decades of Accelerating Genomics

Harisankar Sadasivan, Artur Klauser, Juergen Hench et al. IEEE HPEC.

mm2-gb GPU Accelerated
Workshop 2024

GPU Accelerated Minimap2 for Long Read DNA Mapping

Juechu Dong, Xueshen Liu, Harisankar Sadasivan et al. Biosys Workshop @ ASPLOS 2024.

DTW GPU
Journal 2024

Dynamic Time Warping on GPU for Selective Nanopore Sequencing

Harisankar Sadasivan, Daniel Stiffler, Ajay Tirumala et al. Journal of Biotechnology and Biomedicine.

Minimap2 GPU
Journal 2023

Minimap2 for Accurate Long Read Alignment on GPUs

Harisankar Sadasivan, Milos Maric, Eric Dawson et al. Journal of Biotechnology and Biomedicine.

SquiggleFilter
Conference 2021

An Accelerator for Portable Virus Detection

Presented at MICRO 2021. MICRO Top Picks 2022 Honorable Mention.

RawMap
Journal 2023

Rapid Real-time Squiggle Classification for Read until using RawMap

Harisankar Sadasivan, Jack Wadden, Kush Goliya et al. Archives of Clinical and Biomedical Research.

Press Release

Latest updates.

Social Feed

Latest updates from social media.