MLOps Engineer (AWS & SageMaker)

About the role

The role designs, builds and operates secure, scalable and governed MLOps capabilities that enable reliable deployment and lifecycle management of AI/ML models. It covers production-grade ML pipelines, cloud infrastructure, automation, observability and operational controls, with a primary focus on AWS. It works closely with Data Scientists, Data Engineers, Platform Engineering, Information Security, Model Risk and business stakeholders to accelerate model delivery while maintaining resilience, traceability, cost efficiency and regulatory compliance.

Key responsibilities

• Design and implement scalable MLOps architectures, reusable deployment patterns and standard engineering components across development, testing and production, primarily on AWS.
• Build and maintain automated workflows for data preparation, training, validation, model registration, approval, deployment and retraining using tools such as Amazon SageMaker Pipelines, Step Functions, Airflow or equivalents.
• Implement CI/CD for ML applications and infrastructure using Git-based workflows, automated testing, artifact management and IaC tools such as Terraform or AWS CloudFormation.
• Deploy and operate batch and real-time inference services, ensuring versioning, rollback, scalability and high availability.
• Set up monitoring for model quality, data drift, service health, latency, throughput and failures, with alerts, dashboards, incident response and runbooks.
• Use AWS services such as SageMaker, S3, ECR, ECS/EKS, Lambda, Step Functions, CloudWatch, IAM, KMS and Secrets Manager with secure networking and least-privilege access.
• Configure and integrate Dataiku where applicable, supporting governed deployment, automation and model promotion across environments.
• Maintain model lineage, reproducibility, approvals, audit trails and documentation, and ensure compliance with data privacy, information security, Responsible AI and model risk requirements.
• Optimise compute, storage and inference for performance, resilience and cost, with tagging, usage monitoring and FinOps measures.
• Partner with Data Scientists, Data Engineers and application teams to remove productionisation blockers and improve engineering standards.

Required profile

• At least 7+ years of relevant experience in MLOps, ML engineering, DevOps, cloud engineering, platform engineering or production software engineering, including hands-on deployment and operation of ML solutions in production.
• Proven experience building production-grade ML pipelines, CI/CD workflows, Infrastructure as Code, containerised deployments, monitoring and operational support.
• Experience in banking, financial services or another regulated environment is preferred.
• AWS certification (e.g. Machine Learning Engineer, DevOps Engineer or Solutions Architect) is desirable; Dataiku and Kubernetes/Terraform certifications are an advantage.

Required skills

• Advanced AWS and Amazon SageMaker engineering
• Python and shell scripting; working knowledge of SQL
• CI/CD with GitHub Actions, GitLab CI, Jenkins, AWS Code Pipeline or similar
• Infrastructure as Code with Terraform or AWS CloudFormation, Docker and Kubernetes (ECR, EKS/ECS)
• Model, data and service monitoring, drift detection and observability
• Dataiku integration with AWS Cloud security, IAM, encryption, secrets management and secure networking
• Model governance, lineage, reproducibility, auditability and documentation