All Stories

  1. Reducing the GPU Memory Bottleneck with Lossless Compression for ML

    Article • April 2026, ACM (Association for Computing Machinery)

    Aditya Kamath

  2. POD-A ttention : Unlocking Full Prefill-Decode Overlap for Faster LLM Inference

    Article • March 2025, ACM (Association for Computing Machinery)

    Aditya Kamath

  3. Memory-controller-based lazy memcopy

    Article • June 2024, Institute of Electrical & Electronics Engineers (IEEE)

    Aditya Kamath

  4. Scalable, Programmable and Dense: The HammerBlade Open-Source RISC-V Manycore

    Article • June 2024, Institute of Electrical & Electronics Engineers (IEEE)

    Aditya Kamath

  5. Improved data persistence for GPU + NVM systems

    Article • January 2023, ACM (Association for Computing Machinery)

    Shweta Pandey, Aditya Kamath

  6. Direct data persistence from the GPU through NVM

    Article • February 2022, ACM (Association for Computing Machinery)

    Shweta Pandey, Aditya Kamath

  7. Low-latency, advanced GPU race detector

    Article • October 2021, ACM (Association for Computing Machinery)

    Aditya Kamath

  8. Hardware-based GPU race detector

    Article • May 2020, Institute of Electrical & Electronics Engineers (IEEE)

    Aditya Kamath