Shiva Merla
Principal Software Engineer at NVIDIA
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolShiva Merla is a Principal Software Engineer with eight years focused on cloud-native infrastructure, currently at NVIDIA where he helps enterprises deploy and scale production-grade LLM and generative AI workloads on Kubernetes. He’s a hands-on contributor to flagship open-source projects like the NVIDIA GPU Operator and dcgm-exporter, driving GPU lifecycle automation, node labeling, and Prometheus metrics enhancements that improve GPU utilization for AI/ML. Prior to NVIDIA he built deep storage and multipathing expertise across HPE, NetApp, Nimble and Dell EMC—developing CSI drivers, connection managers, and Linux failover drivers that underpin resilient enterprise storage. Shiva combines low-level Linux and storage kernel knowledge with modern DevOps and operator patterns, uniquely bridging device drivers to cloud-native orchestration. Based in San Jose, he’s known for pragmatic automation that shortens bring-up time while maximizing hardware efficiency for large-scale AI deployments.
8 years of coding experience
10 years of employment as a software developer
BE, Computer Science, BE, Computer Science at Sri Jaya Chamarajendra College Of Engineering. MysoreBachelor of Engineering (B.E.), Computer Science, Bachelor of Engineering (B.E.), Computer Science at Sri Jayachamarajendra College of Engineering, Mysore, India
English, Hindi, Telugu, Kannada
Github Skills (18)
kubernetes10
docker10
containerization10
dockers10
system-monitoring10
go10
kubernetes-pods10
devops10
cicd9
helm9
openshift9
prometheus9
crt8
yaml8
gpu8
Programming languages (12)
Github contributions (5)
NVIDIA/gpu-operator
Sep 2020 - Jan 2023
NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes
Role in this project:
Backend & DevOps Engineer
Backend & DevOps EngineerContributions:14 releases, 40 reviews, 627 commits in 2 years 4 months
Contributions summary:Shiva's contributions focused on the NVIDIA GPU Operator, primarily related to enhancing its capabilities and streamlining deployments within Kubernetes environments. The user implemented features for automated GPU node labeling and improved driver container management, addressing specific issues with the Node Feature Discovery (NFD) labels. The user also updated the operator to deploy GFD, added configurations to customize components of the driver, enabled the creation of operator components, and streamlined the Operator by optimizing the daemonset code. The user was responsible for adding and updating the OLM bundle, CRD, and CSV updates for OLM support
gpukubernetesnvidiaoperatorcuda
NVIDIA/dcgm-exporter
Oct 2020 - Jan 2023
NVIDIA GPU metrics exporter for Prometheus leveraging DCGM
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:11 commits, 1 PR, 1 comment in 2 years 3 months
Contributions summary:Shiva's contributions primarily focused on enhancing the functionality and configuration of the dcgm-exporter. They implemented features to check for and collect DCP metrics, modifying several Go files to enable this capability. Furthermore, they integrated the use of the `NODE_NAME` environment variable for metrics, demonstrating a DevOps approach to containerized application configuration. Additionally, they addressed whitespace issues and added a VGPU license status to the default watch list, improving code quality and functionality.
gpunvidiaprometheus
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.