Pramod Ramarao
Software Engineer at Sift
California, United States
Summary
🤩
Rockstar🎓
Top SchoolPramod Ramarao is a seasoned software engineer with 10+ years designing high-performance data platforms and streaming infrastructure, now building real-time systems in Rust at Sift. He brings deep systems and observability expertise from senior roles at Google—leading multi-petabyte cloud migrations, architecting sub-second real-time pipelines, and driving $8M+/month GCP adoption for major enterprises. Earlier work bootstrapping enterprise data lakes, Teradata offloads, and near-real-time analytics reflects a rare mix of hands-on systems engineering and strategic technical leadership. An active open-source contributor and DevOps architect, he has improved NVIDIA GPU monitoring tooling (dcgm-exporter) for containerized environments and added remote host engine support. Based in California, he pairs low-level Unix/storage/automation roots with modern cloud-native practices, and mentors teams to adopt observability-first, production-hardened patterns. Notably, his background spans both hardware-level recovery automation and large-scale cloud data migrations, giving him a pragmatic edge when bridging legacy infra and cloud-native platforms.
10 years of coding experience
8 years of employment as a software developer
Bengaluru UniversityDiploma, Electronics and Communication Engineering, Diploma, Electronics and Communication Engineering at Sri Jayachamarajendra PolytechnicJnana Mitra High School
English, Kannada, Hindi
Github Skills (16)
system-administration10
monitoring10
sh10
nvidia10
kubernetes10
script10
docker10
go10
prometheus10
shell10
gpu10
containerization10
dockers10
scripting10
cicd10
Programming languages (5)
Github contributions (5)
NVIDIA/gpu-monitoring-tools
Sep 2020 - Jul 2021
Tools for monitoring NVIDIA GPUs on Linux
Role in this project:
DevOps Engineer
DevOps EngineerContributions:13 commits, 2 PRs, 12 comments in 10 months
Contributions summary:Pramod implemented changes to the `dcgm-exporter` entrypoint script to include error handling for the `setcap` command, addressing issues in OpenShift environments. They updated the DCGM version and headers, incorporated changes to the default interval, and added the ability for `dcgm-exporter` to connect to a remote host engine. These changes reflect a focus on improving the deployment and configuration of the monitoring tools within containerized environments. Further work included the merging of branches which introduced updates to DCGM and the addition of a GPU card name to the labels.
cudanvidiagpulinuxgpus
NVIDIA/dcgm-exporter
Sep 2020 - Jul 2021
NVIDIA GPU metrics exporter for Prometheus leveraging DCGM
Role in this project:DevOps Engineer & System Architect
DevOps Engineer & System ArchitectContributions:11 commits in 10 months
Contributions summary:Pramod primarily focused on setting up and configuring the `dcgm-exporter` project, which is a NVIDIA GPU metrics exporter for Prometheus. Their contributions included modifying the entrypoint script to add the `cap_sys_admin` capability, improving the Docker build process. They also integrated updates from other branches, including updated DCGM versions and adding support for remote host engine connections, which demonstrates a focus on infrastructure, monitoring, and system architecture.
exporternvidialeveraginggpumetrics-exporter
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.