Hela Souguir

• CentOS Systems Administration: Installation and administration of a fleet of over 200 VMs on VMware vSphere, managing various services such as MariaDB, HAProxy, Kubernetes, Docker, Vault, ELK, Keycloak, TYK.

• Kubernetes: On-premises infrastructure supervised via Rancher. Deployments via ArgoCD, storage with Longhorn, Traefik reverse-proxy, cert-manager + Cloudflare, multi-environments via Kustomize, registries via Harbor.

• Docker: Setup of multi-container applications with Docker Compose.

• Ansible: Complete VM provisioning, automated deployment of stacks (K8s, Vault, ELK, HAProxy...), configuration, and remediation.

• Linux Security: System hardening, automated CIS remediations with CSV reports, PAM, sudo, SSH, auditd policies.

• Databases: Containerization, configuration, and management of MariaDB. Intelligent routing between clients/applications and MariaDB instances via ProxySQL. Administration of PostgreSQL pods. Application load balancing of Adminer via HAProxy.

• Monitoring: Multi-level monitoring via Prometheus, Grafana, Telegraf, InfluxDB, VictoriaMetrics, Cluster.

• VMware:

• Apache httpd (reverse proxy): Setup and configuration of an Apache httpd reverse proxy with high availability support via KeepAlived (VIP, automatic failover).

• Bitbucket & Jenkins: Setup of CI/CD pipelines.

• Backups: Veeam and MinIO for VMs and containerized services.

• Contribution to the System Team of the Orange France Conversational Interfaces Safe train, the project includes chatbots dedicated to Orange public customers and internal employees.

○ Strengthening DevOps infrastructure for public chatbots, improving customer engagement on the Orange France customer portal and support.

○ Migration to GKE: Contribution to the migration of the chatbot architecture to Google Kubernetes Engine, boosting scalability and resilience.

○ IaC with Terraform: Automated deployment of cloud infrastructure.

○ Cluster Administration: Optimized management of Kubernetes and Docker Swarm clusters.

○ Production Deployment and Incident Management: Active participation in production deployments and rapid incident resolution, ensuring increased service responsiveness and reliability.

○ Autoscaling Improvement: Optimization of microservice deployment workflows using Kubernetes, with a focus on advanced autoscaling.

○ Cost Optimization: Application of FinOps principles to optimize costs on GCP.

○ Advanced Monitoring: Development and implementation of monitoring and alerting systems using Cloud Monitoring, Grafana, and Prometheus, complemented by Tempo for operation tracing.

○ Management and configuration of BRMC components (Orange private cloud): Efficient administration of bots on VMs hosted on BRMC before migration to GCP.

○ Configuration Management: Use of Ansible for precise configuration management and automated deployment of the technical stack used.

○ Security and Compliance: Integration of code analysis and vulnerability tools (Checkmarx, Sonar, Trivy, Gitleaks, Dependency-Check, …) and use of Google Cloud Armor for advanced protection against threats and network attacks on the Kubernetes cluster.

○ Automation and Continuous Integration: Setup of CI/CD pipelines with GitLab CI.

• Development and optimization of DevOps processes for ORANGE's Scribe application, a centralized correspondence management solution for Orange:

○ CI/CD Pipelines: Design and optimization of continuous integration and continuous deployment (CI/CD) pipelines for Scribe's user interfaces, using GitLab CI to automate workflows and ensure fast, reliable deliveries.

○ Code Quality and Security Analysis: Adoption of SonarQube, OWASP ZAP, and Checkmarx for in-depth analysis and vulnerability scanning, integrated into our CI/CD pipelines.

○ Deployment Infrastructure and Configuration Automation: Setup of deployment architectures on virtual machines hosted on Orange's BRMC and development of Ansible roles and playbooks for efficient automated deployment of multiple sub-projects of the Scribe application on non-production and production servers. This approach involved automated configuration of the entire technical stack required for the application's proper functioning, including services, databases, and inter-service communication tools. This strategy facilitated smoother and faster production deployments, and also ensured the stability and performance of deployment environments.

○ Automation and Testing: Implementation of systematic unit and functional tests, using frameworks like JUnit, Protractor, and Robotframework, to validate the logic of individual components and the integrity of application flows.

○ DevOps Support: Supporting the Scribe team to facilitate the integration and adoption of DevOps practices. This support encouraged automation and significantly improved the efficiency and consistency of development and deployment processes within the team.