Rishi Madduri

Rishi is a software engineer on the AIOSS Ecosystems team. His expertise lies in C++ and GPU programming. He is currently focused on developing and optimizing HIP kernels to accelerate machine learning workloads on the CDNA architecture.

Posts by Rishi Madduri

Enabling FlashInfer on ROCm for Accelerated LLM Serving

FlashInfer is an open-source library for accelerating LLM serving that is now supported by ROCm.

October 01, 2025 by Rishi Madduri, Diptorup Deb, Debasis Mandal, Clint Greene, Mukhil Azhagan Mallaiyan Sathiaseelan, Yao Liu, Phani Vaddadi, Vish Vadlamani

Efficient MoE training on AMD ROCm: How-to use Megablocks on AMD GPUs

Learn how to use Megablocks to pre-train GPT2 Mixture of Experts (MoE) model, helping you scale your deep learning models effectiveness on AMD GPUs using ROCm

March 23, 2025 by Fabricio Flores, Rishi Madduri, Yao Liu, Phani Vaddadi, Vish Vadlamani