Srinivas Kotapati
Software Development Manager BIOS at Amazon Web Services (AWS)
San Ramon, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSrinivas Kotapati is a seasoned firmware and platform engineering leader with over two decades of experience building BIOS/UEFI and silicon bring‑up for Intel, AMD, Apple, Microsoft, and AWS. He has driven major architectural transformations—most recently leading Azure’s migration from AMI BIOS to an in‑house UEFI framework and now heading BIOS engineering for next‑generation AWS hardware—while reducing cross‑architecture enablement time from months to days. Known for blending deep low‑level technical expertise (CPU init, memory training, PCIe, ACPI, secure boot) with strong program management, he created standardized CI/CD, shift‑left validation, and cross‑vendor delivery models that cut late defects and technical debt. At Apple he helped bridge disparate firmware cultures during transitions from Intel to T2 and early Apple Silicon, scaling teams and training programs through complex platform shifts. An active contributor to the PyTorch MPS backend, he also brings hands‑on open‑source experience improving GPU/Metal support and CI stability. Based in San Ramon, he combines a systems‑level perspective with pragmatic execution to deliver reliable firmware at hyperscale.
11 years of coding experience
25 years of employment as a software developer
Bachelor of Technology (BTech) ECE, Bachelor of Technology (BTech) ECE at Andhra UniversityM.Tech, M.Tech at R.E.C, Kurukshetra
Github Skills (10)
pytorch10
machine-learning10
gpu10
python10
mps10
metal-performance-shaders10
neural-network9
deep-learning9
autograd8
testing7
Programming languages (3)
Github contributions (5)
pytorch/pytorch
Apr 2024 - Feb 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:16 reviews, 17 PRs, 1 push in 10 months
Contributions summary:Srinivas made several contributions related to the Metal Performance Shaders (MPS) backend for PyTorch. This included adding MacOS 15 runners for testing, implementing workarounds for MPS operations, and enabling additional tests for MPS CI runs. They also addressed synchronization issues within the MPS framework. The work focused on improving MPS support and stability.
pythongpu-accelerationdeep-learninggpunumpy
skotapati/MobileTPS
Mar 2016 - Mar 2016
Contributions:10 commits, 6 pushes, 1 branch in 14 days
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.