Currently working on the backend and the infrastructure of https://exalyze.io. Developped the frontend (HTML, Bootstrap) and a large part of the backend application backend logic. I also lead the effort on automation (using Gitlab CI, Docker Swarm and Ansible) making as easy as possible to deploy the application. I also greatly improved the platform observability by setting up systematic application monitoring using Prometheus, Loki and Grafana.

As part of the VOI Platform team, I am in charge of maintaining the platform and tools supporting VOI, from the core infrastructure on GCP, to the CI/CD pipeline (GitHub Actions and ArgoCD), through monitoring (Prometheus, Grafana and Mimir). In less than a year, I have achieved amongst other things the successful zero downtime migration of our workload to separate Kubernetes namespaces and the rollout of Grafana Mimir as our metric storage backend. My proficiency with Go and Kubernetes allowed me to make meaningful contributions across our application stack and to be recognized as a technical reference in our organization. This is also a very productive period for me as an open source developer as my position inspired me many side projects like kudo, a Kubernetes controller that allows to implement secure permission escalation, and kxds, a control plane for gRPC xDS service discovery. I also won the second prize of the Traefik 3.0 Haekathon with my work integrating SPIFFE with Traefik. Last but not least, I implemented an highly available solution to push metrics to Mimir using Prometheus Agents and Kubernetes Leader Election which got very good feedback from the Prometheus maintainers. For more information see the project section.

While still being involved on a daily basis at a technical level, this position was also the first time I assumed a leadership role in a team. On the technical side, I wrote amongst other things the support for Instagram stories, an ETL pipeline for email analysis, performed various infrastructure updates and contributed to major new features of the platform. As a team leader, I successfully set up a development workflow from the feature inception to production. This contribution greatly improved the team productivity while ensuring an acceptable workload and provides a better visibility to other stakeholders of the company.

For a couple of years, with @DamienDuportal, we have been giving a 3 days lecture to a MSc level group of students related to CI and CD, tests and “good” practices of software development. This lecture aims to be the more practical possible and relies on SaaS tools like gitpod for students development environments and GitHub actions as CI/CD engine to ensure that. You can find the slides here, (only in French).

Involved in the Traefik Enterprise team, I contributed to numerous features embedded in the product. I proposed and implemented many improvements on it and impacted positively its reliability. For instance setting up an end to end testing framework or the systematic writing of unit tests. I was also very involved with the support team and regularly participated to client facing calls. Traefik Labs was my first experience in a truly distributed and multicultural company where clear communication and inclusion are a must, and I have to say that I loved this kind of environment.

Hired initially as backend developer, my previous experience with server management quickly got me involved with the operations as well. I was put in charge of important subjects like moving our ecommerce websites to HTTPS, rewrote, extended several applications and deployed a whole new monitoring system based on Datadog. I also had the opportunity to teach an intern during my time and I found it to be a rewarding experience.

As the company’s first employee, I was involved in a small distributed team between Lyon and San Francisco. We built from scratch multiple products centered around content marketing and native advertising and it finally lead the Upfluence software to be one of the most powerful influencer search engine available. My position was fullstack developer and I worked on a complex application stack involving numerous microservices written in 4 major different languages, golang included. This was also my first experience with production and gave me the opportunity to learn cutting edge subjects like containers, orchestration and distributed systems. I’m proud to see that many years later, Upfluence is a successful company and some of my code is still running at its core

Hired as developer on a complex application built to refund clients in case of bank bankruptcy. It involved complex business rules and being able to handle an important amount of open cases. This was my first experience as web developer, in a large team and in a large company. I gave various talks about test driven development, like this presentation as I understood how important it is on projects of this scale.

I had two primary missions in that position: First was to build middleware and user interface application for a “second screen” set top box system, using DLNA, linux and web technologies. Second was to build a multiplatform desktop application dedicated to control Awox Striim series devices. It featured a DLNA control point, and also custom features like devices zoning and lighting control over network.