Romain Cledat
Member Of Technical Staff at Anthropic
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolRomain Cledat is a Senior Software Engineer based in San Francisco with 11 years of experience building robust back-end and DevOps solutions. At Netflix since 2019, he focuses on runtime reliability for AI/ML platforms, improving task termination, log flushing, and environment management in the widely used open-source Metaflow project. He brings practical expertise in process lifecycle, batch execution clients, and Conda/environment variable integrations that make large-scale ML workflows more resilient. Known for pragmatic fixes that strengthen testing and operational behavior, he balances deep systems thinking with hands-on code changes. Colleagues rely on him to turn brittle shutdown scenarios into predictable, debuggable behavior in production pipelines.
11 years of coding experience
21 years of employment as a software developer
Preparatory School, Preparatory School at Lycee du ParcPhD CS, PhD CS at Georgia Institute of TechnologyMS General Engineering, MS General Engineering at Centrale LyonCite Scolaire Internationale
French, English
Github Skills (4)
process-management10
python10
devops9
parallel-processing8
Programming languages (9)
Github contributions (5)
Netflix/metaflow
Dec 2019 - Jan 2023
Build, Manage and Deploy AI/ML Systems
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:45 releases, 841 reviews, 383 commits in 3 years 1 month
Contributions summary:Romain primarily worked on improving the runtime environment of the Metaflow framework by addressing issues related to process termination and shutdown procedures. Their contributions involved modifying the `runtime.py` and `batch_client.py` files to enhance the robustness of task termination, including handling forked processes and ensuring proper log flushing. The user also incorporated several changes that improved testing and added features. They also made changes related to Conda environments and environment variable configuration.
aimachine-learningmodel-managementmlml-platform
romain-intel/conda-lock
Dec 2022 - Dec 2025
Lightweight lockfile for conda environments
Contributions:4 reviews, 1 PR, 45 pushes in 3 years 1 month
condalockfile
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.