Jordan Jacobelli
DevOps Engineer at NVIDIA
France
Summary
🤩
Rockstar🎓
Top SchoolJordan Jacobelli is a DevOps engineer with 11 years of experience focused on CI/CD, build and release automation, and GPU-accelerated software workflows, currently driving DevOps at NVIDIA from France. He has a strong track record optimizing pipelines for prominent RAPIDS projects (cuDF, cuML, cuGraph, cuSignal, cuSpatial, RMM, RAFT), improving build speed with Ninja and mambabuild, modernizing conda environments, and automating package uploads and testing. Jordan’s work spans both cloud-native GPU operator deployments in Kubernetes and low-level build orchestration for CUDA-accelerated ML libraries, demonstrating fluency across infrastructure and ML engineering domains. Former SRE and container software roles (including an NVIDIA internship and Smart AdServer) give him practical production reliability experience alongside release engineering. He often focuses on non-obvious productivity gains—like replacing slow conda flows, removing legacy channels, and converting GPU test jobs to conda—to shave hours off build and test cycles. Educated in France (EPITA and IUT Aix-Marseille), he combines academic grounding with hands-on open-source contributions to widely used GPU tooling.
11 years of coding experience
1 year of employment as a software developer
EPITA: Ingénierie InformatiqueIUT Informatique, IUT Informatique at IUT d'Aix Marseille
French, English
Github Skills (38)
e2e-testing10
kubernetes10
github-ci10
docker10
python10
e2e-test10
bash10
conda10
cmake10
dockers10
cicd10
nvidia10
automation10
rapids10
gpu10
Programming languages (15)
Github contributions (5)
rapidsai/cuspatial
Feb 2021 - Jun 2022
CUDA-accelerated GIS and spatiotemporal algorithms
Role in this project:
DevOps Engineer
DevOps EngineerContributions:8 reviews, 19 commits, 28 PRs in 1 year 4 months
Contributions summary:Jordan primarily focused on improving the continuous integration and continuous delivery (CI/CD) pipeline for the repository. Their contributions included setting up builds with CMake, integrating Ninja for faster builds, and configuring the upload of conda packages. They also addressed issues with conda channels, Python versions, and build script configurations to optimize the CI/CD workflow. Furthermore, the user updated test scripts to align with changes in dependencies.
cudagpu-programminggpu-accelerationgpuaccelerated
rapidsai/rmm
Feb 2021 - Oct 2022
RAPIDS Memory Manager
Role in this project:DevOps Engineer
DevOps EngineerContributions:28 reviews, 13 commits, 33 PRs in 1 year 8 months
Contributions summary:Jordan focused on improving the CI/CD pipeline and build processes for the `rmm` repository. Their contributions included modifying build scripts to remove nightly conda channels and to prepare for Python 3.7 removal. The user also implemented the use of `mambabuild` for faster conda package builds and converted GPU test jobs to use conda. Furthermore, they integrated GitHub Actions workflows for automated builds, tests, and package uploads.
cudamemory-managementmemorycpppython
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.