Freelancer profile translated to English.
Back to original languageAbout Arthur
I am an independent DevOps consultant. I help teams migrate to Kubernetes, reduce their cloud costs, and implement infrastructure that holds up in production.
What I do specifically:
- Migrating applications to Kubernetes (Scaleway, AWS, on-premise)
- Reducing infrastructure costs (up to -72% in real cases)
- Implementing IaC, CI/CD, monitoring, alerting
- Training and coaching teams on Kubernetes, Helm, GitOps
A recent example: a client had started their Kubernetes migration on their own. Pods restarting in loops, 300,000 lines of IaC, lost teams. In 6 months, I reduced the IaC to 8,000 lines, stabilized production, and trained the teams on best practices.
10 years of experience. Clusters from 3 to 500 nodes. From startups to large accounts.
Need an external perspective on your infrastructure? Write to me.
French
Native or bilingual
English
Native or bilingual
Remote only
Primarily works remotely
Experience
- Conservatoire AugmentéInfrastructure ManagerEDUCATION AND E-LEARNINGJuly 2025 - Today (1 year and 1 month)Lyon, FranceMission:Design and deploy the cloud infrastructure for an LMS for the Paris Conservatory.Full responsibility for the infrastructure of an e-learning platform (LMS) for a music conservatory, from technical scoping to production deployment.What I delivered:
- Infrastructure as Code: integral design with OpenTofu on Scaleway — multi-zone Kubernetes cluster, VPC with private gateway, autoscaling up to 30 nodes
- Application Architecture: containerization of WordPress with Docker, deployment via ArgoCD in GitOps mode, Valkey cache, NFS storage for uploads
- Centralized Authentication: implementation of Keycloak (SSO/OIDC) to unify access between services
- High Availability Databases: migration of MySQL databases to Percona XtraDB (PXC) clusters with synchronous replication, automated backups, and data integrity validation
- Cost Optimization: complete infrastructure maintained below €1,000/month through instance sizing based on actual CPU/RAM ratios and migration of managed databases to PXC clusters hosted on the Kubernetes cluster
- Performance: selection of high CPU/RAM instances (HC), application caching with Valkey, synchronous replication databases to minimize read latency
- Monitoring & Alerting: VictoriaMetrics stack with 1-year retention, Grafana dashboards (MySQL, Galera, cluster), alerts via Microsoft Teams webhook and Uptime Kuma
- Security: encryption of secrets with SOPS/age, Scaleway Secret Manager, External Secrets Operator for automatic synchronization within the cluster
- Multi-environments: dev, staging, and production — each environment isolated and reproducible
Stack:Kubernetes (Scaleway Kapsule) · OpenTofu · ArgoCD · Docker · WordPress · Keycloak · Percona XtraDB · Valkey · NFS · VictoriaMetrics · Grafana · SOPS · External Secrets Operator - Exxact RoboticsFreelance DevOps Cloud AWSTECHFebruary 2025 - Today (1 year and 6 months)Lyon, FranceMission:Complete reverse engineering and migration of an agricultural robotics IoT platform on AWSExxact Robotics (EXEL Industries group), a platform for piloting autonomous agricultural robots, mission planning, and fleet management. The existing infrastructure, deployed without documentation, required complete reverse engineering before migration.What I delivered:
- Reverse Engineering: audit and complete mapping of an AWS infrastructure in production without documentation — inventory of 27+ services, dependencies, data flows, and configurations
- Multi-Account AWS Migration: complete replication to a new Exxact account — 18 Lambdas (.NET/Python/Node.js), API Gateway, Cognito, Aurora MySQL Serverless v2, DynamoDB, ElastiCache Redis, S3, CloudFront
- Multi-Account & Resilience Refactoring: restructuring the architecture to support a multi-account AWS model, strengthening application and infrastructure resilience
- Infrastructure as Code: creation of code for the entire system in OpenTofu modules with workspaces to isolate 3 environments (prod, rc, beta) across multiple AWS accounts
- CI/CD: creation of deployment pipelines to automate infrastructure provisioning and application deployment
- IoT: implementation of AWS IoT Core — MQTT, X.509 certificates, policies, management of Things for machine-to-cloud connectivity of 40+ robots
- Containerization: multi-architecture Docker (ARM64 for robots, x86 for cloud), systemd configuration for dual-service deployment on embedded hardware
- Data Migration: RDS transfer between accounts, DynamoDB/IoT/RDS synchronization, network conflict diagnosis for multi-service deployment on robots
Stack:AWS (Lambda · API Gateway · Aurora MySQL · IoT Core · Cognito · DynamoDB · ElastiCache · S3 · CloudFront · SNS · EventBridge · Route53) · OpenTofu · Docker · Azure DevOps · .NET 6 · MQTT - Enedis - LyonKubernetes ExpertENERGY AND UTILITIESDecember 2023 - November 2025 (1 year and 11 months)Lyon, FranceMission:Kubernetes expertise and reliability of Enedis' internal containerized platform.Enedis, the French electricity distribution network operator, operates a shared Kubernetes offering for its internal business projects. Expertise mission to ensure production reliability, support application teams, and contribute to the design of the next generation of the offering.What I delivered:
- Alerting and Incident Response Redesign: rethinking the alert pipeline (rules, routing, severities) and response processes to reduce noise and accelerate resolution
- Self-Remediation: implementation of self-remediation software triggered by Prometheus alerts, to automatically resolve recurring incidents without human intervention
- Removal of Resource Allocation "Bell": removal of a custom Python script that blocked rolling updates as long as pod requests were not at 95% of actual consumption. This constraint pushed projects to declare absurdly low requests to pass the check, which degraded application performance and node resilience. The removal allowed for healthy sizing and unblocked deployments.
- Project Support: coaching application teams — Kubernetes onboarding, deployment best practices, resolving production issues
- Architecture: proposal for infrastructure architecture redesign to overcome identified limitations
- Observability: creation of production-quality Grafana dashboards for platform and business workload monitoring
- Offering Evolution: contribution to technical discussions and challenges for V2 of the team's Kubernetes offering (architecture, service model, user journey)
Stack:Kubernetes · Rancher · VictoriaMetrics · Grafana · Traefik · GitLab CI
Reviews
Recommendations
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Ingénieur en Informatique, Technologies de l''informationexia.CESI : Ecole d'Ingénieurs Informatique2018Ingénieur en Informatique, Technologies de l''information
Certifications
- Certified Kubernetes AdministratorLinux Foundation2021
- Certified Kubernetes Security SpecialistLinux Foundation2022