Hi, I'm Louis Labeyrie 👋
DevOps engineer & SRE who loves ops so much that one day I will host the world in my garage.
About
I fell into DevOps the day I realized I'd rather spend three hours automating a task than ten minutes doing it by hand. Freshly graduated from Polytech Montpellier with an engineering degree in DevOps, I spent three years at PixelsCoder building cloud workspaces with Kubernetes and Terraform, wiring up CI/CD pipelines that cut deployment effort by 70%, and containerizing anything that stood still long enough.
Most recently, at Elevait in Dresden, I designed an MLOps platform on Kubernetes: GitOps deployments with Flux, automated training and hyperparameter tuning, dataset versioning, model serving and drift detection, all watched over by Grafana.
My happy place is somewhere between a Helm chart and a Grafana dashboard, turning fragile, manual processes into infrastructure that just works. I've also caught the open-source bug, contributing CI improvements to projects like Deskflow and database resilience patches to useSend.
The long-term plan? A garage full of servers, a domain for every side project, and the conviction that if it can be automated, it should be.
Work Experience
- Benchmarked MLOps tools against the company's needs and designed an ML platform on Kubernetes.
- Investigated and resolved incidents on Kubernetes clusters, and deployed applications following GitOps practices with Flux.
- Built an end-to-end proof of concept on a development cluster: automated training with hyperparameter tuning, dataset versioning with DVC, model serving, drift detection (Alibi Detect) and Grafana monitoring.
- Trained future users to move from manual, ad-hoc training workflows to a tracked and reproducible process.
- Developed a cloud workspace solution using Kubernetes and Terraform, reducing setup time by 50%.
- Containerized and orchestrated internal services using Docker and Helm on a Kubernetes cluster.
- Automated CI/CD workflows with GitHub, reducing manual deployment steps by up to 70%.
- Optimized infrastructure and cloud services to enhance uptime and reduce recovery time.
- Developed an incident management system and automated internal IT tasks.
- Deployed and secured Linux servers, implemented a SIEM system (Suricata) for network monitoring.
- Managed user accounts and IT equipment on Windows servers, providing support and resolving technical issues.
Education
- Strengthened operations skills through substantial projects.
- Explored different cloud providers (GCP, AWS, OpenStack).
- Worked on long-term group projects, including a reproduction of the Discord social app.
- Gained strong foundational skills in statistics and data analysis.
- Learned Python and SQL for data manipulation and analysis.
Skills
Projects
Open Source
ci: add step to install and test package binaries, reusable CI composite action that validates built packages on Linux, macOS, and Windows by installing them and running version checks.
fix: auto-retry database connection after server restart, exponential backoff with jitter for Prisma operations that fail due to transient connection errors.
Update detail to the 2026 edition, updated CI workflows, Firebase config, navigation, hero section, sponsors block, tickets block, and location page for the upcoming edition.
Add CFP (Call for Papers) section, added a link to the call for papers on the hero section of the conference website.
feat: add SetUser API and CLI to update user profile fields, gRPC/REST endpoint and CLI command to update display_name, email, and profile_pic_url with optional field semantics.
Get in Touch
I'm always open to discussing new projects, opportunities, or collaborations. Feel free to reach out!