Oussama ARIFALLAH

Design and automation of CI/CD pipelines for Kubernetes (EKS) and cloud infrastructure using Jenkins and GitFlow. Integrated automated security gates including Static Application Security Testing (SAST) with SonarQube and Checkmarx One to ensure high-quality, secure deployments. Provisioning and management of AWS resources via Terraform; developed Helm Charts to standardize Kubernetes deployments. Implemented systematic security validation of K8s manifests using Kubesec and Kube-linter to prevent misconfigurations prior to deployment. Developed highly scalable serverless solutions leveraging the AWS stack: Lambda, Step Functions, EventBridge, API Gateway, SQS, SNS, and DynamoDB. Implemented SFTP/FTPS servers on EKS with secure integration to AWS S3 via AWS Transfer Family, centralizing the management of business-critical transfers. Designed and executed Blue/Green deployment strategies to achieve Zero Downtime and minimize operational risks during production releases. Technical Lead on microservices hardening, including the implementation of non-root images, Network Policies, and container isolation via AppArmor profiles to restrict Linux kernel capabilities. Orchestrated "Lift and Shift" infrastructure migrations between AWS accounts, involving resource isolation and architectural reorganization. Deployed and configured observability stacks for AWS and Kubernetes using Splunk, CloudWatch (Logs/Metrics), Container Insights, Fluentbit, FluentD, Prometheus, and Grafana. Reduced AWS spend through granular analysis with Cost Explorer and implemented Tagging Policies for internal cost allocation (chargeback/showback). Developed Proof of Concepts (PoCs) to validate architectural solutions before large-scale industrialization on AWS. Automated training, testing, and deployment (MLOps) by applying CI/CD principles to ML models, integrating code quality scans (SonarQube) and application security (Checkmarx One). Collaborated with Data Scientists to industrialize notebooks into production-ready environments using AWS SageMaker and Step Functions. Containerized ML models via Docker, utilizing Checkmarx One for inference code vulnerability scanning and SCA (Software Composition Analysis) for open-source dependencies. Managed images via AWS ECR and Azure ACR for deployment on Amazon ECS. Deployed and interconnected vector databases (AWS Neptune, Azure AI Search) to implement GraphRAG systems on SageMaker and Azure ML. Designed autonomous AI agents using LangChain and orchestrated models via AWS Bedrock, with full infrastructure provisioning via AWS CDK. Monitored LLM consumption (AWS Bedrock) and vector database usage (Neptune) by implementing budget alerts and strict tagging for project-based cost tracking and ML monitoring. Led technical discovery sessions to extract requirements and design bespoke solutions tailored to business needs. Visualized complex business logic and workflows using UML. Managed software development lifecycles and project tracking using the Atlassian suite (Jira & Confluence). Delivered technical demonstrations and project updates to stakeholders and international teams in both French and English.