Rohith Nallamaddi is a senior software engineer with nine years of experience building and optimizing ML infrastructure, model deployments, and inference pipelines across industry leaders including Apple, AWS, and AMD. He specializes in accelerating deep learning workloads on specialized hardware (Trainium, Inferentia, GPUs) and has hands-on expertise in backend and DevOps work for production ML systems, including contributions to the widely used PyTorch Serve project. His background spans signal processing, computer vision, and data science, grounded in an MSc in Electrical and Computer Engineering and practical research in acoustic and speech processing. Rohith combines low-level performance work (C/C++, CUDA) with higher-level ML tooling (TensorFlow, Keras, PyTorch) to move models from research to scalable production. He’s known for pragmatic optimizations—updating build and CI pipelines, integrating libraries into frameworks, and tuning transformer inference—that yield measurable efficiency gains. Based in San Jose, he blends academic rigor with product-focused engineering to solve challenging deployment and latency problems.
9 years of coding experience
9 years of employment as a software developer
Bachelor's Degree, Electrical Engineering, Bachelor's Degree, Electrical Engineering at Indian Institute of Technology, Bhubaneswar
Serve, optimize and scale PyTorch models in production
Role in this project:
DevOps Engineer & Backend Engineer
Contributions:67 reviews, 15 commits, 21 PRs in 3 months
Contributions summary:Rohith primarily focused on updating and maintaining the build and deployment infrastructure for the PyTorch Serve project. They made several modifications to the installation scripts, including updating Java versions on different operating systems and integrating Maven for build management. Additionally, they worked on updating the CI/CD pipeline to use Java 17, which demonstrates an understanding of backend dependencies and build processes. Furthermore, the user contributed to the kserve gpu tests.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.