DevOps and Machine Learning Operations Engineer
Location: Manchester, UKContract Type: Contract Position
About the Project
Join a specialist software development team delivering a new, business-critical technology platform for an established international organisation. This greenfield project involves modern cloud architecture, data-intensive applications, and AI-enabled capabilities. You will collaborate with a multidisciplinary team, taking the platform from development through to production in a live business environment where security, scalability, maintainability, data protection, and production readiness are essential.
Role Purpose
As an Azure DevOps and Machine Learning Operations Engineer, you will design, automate, and operate Microsoft Azure environments for production software and machine-learning services. Your responsibilities will include infrastructure as code, delivery pipelines, secure networking, and operational support.
Applicants must have experience operating deployed systems, investigating incidents, testing recovery, and managing model registries, evaluation automation, versioned data and model deployments, and controlled rollback.
Main Responsibilities
- Microsoft Azure Platform Architecture: Design and operate Azure environments using services such as Container Apps, Container Registry, Azure Database for PostgreSQL, Blob Storage, and Service Bus. Configure Front Door, Web Application Firewall, API Management, Microsoft Entra ID, and Key Vault. Select and manage container orchestration solutions, support capacity decisions, and understand network isolation impacts.
- Infrastructure as Code: Create reusable Terraform modules for environments, networking, container runtimes, data services, and monitoring. Manage remote state, secrets, and configuration drift. Ensure infrastructure changes are versioned, tested, and deployed through controlled pipelines.
- Continuous Integration and Delivery: Build GitHub Actions or Azure DevOps pipelines for TypeScript and Python services. Implement automated tests, scanning, secret detection, and infrastructure validation. Manage environment promotion, deployment approvals, and database migrations with documented recovery paths. Plan for controlled releases and rollbacks.
- Artificial-Intelligence Delivery Pipelines: Automate dataset versioning, model registration, and deployment of online or batch endpoints. Preserve model and environment versions, support shadow evaluation, controlled releases, model suspension, and fallback. Monitor embedding refreshes and search or graph rebuilds.
- Platform Security: Implement least-privilege access, managed identities, multi-factor authentication, and controlled administration. Integrate Key Vault, manage secret and certificate rotation, encryption, and network isolation. Use security monitoring and enforce restricted access as needed.
- Observability and Operations: Establish distributed tracing, metrics, and logs using OpenTelemetry and Azure Monitor. Monitor service health, model performance, provider costs, and ensure actionable alerts and runbooks are in place. Define service-level indicators and objectives, and conduct incident reviews.
- Reliability and Recovery: Implement and test backups, database restoration, storage lifecycle controls, and index rebuilds. Develop procedures for interrupted processing and degraded operation. Coordinate load and resilience tests, and verify release readiness.
- Cost and Capacity Management: Measure and manage compute, database, storage, network, and telemetry consumption. Set budgets, investigate growth, and use autoscaling and scheduled shutdowns. Ensure cost reductions do not compromise performance or security.
Requirements
- At least six years of DevOps, cloud-platform, or site-reliability experience, with substantial production responsibility on Microsoft Azure.
- Strong skills in Terraform, Docker, and CI/CD, with practical delivery on Azure Container Apps or Azure Kubernetes Service.
- Experience with secure virtual networks, private endpoints, Microsoft Entra ID, Key Vault, Azure Service Bus, Blob Storage, and Azure Database for PostgreSQL.
- Ability to troubleshoot connectivity, identity, deployment, and capacity issues.
- Experience supporting Node.js and Python services, distributed tracing, vulnerability remediation, and tested disaster recovery.
- Proven incident management skills, including service restoration and implementing corrective actions.
- Experience with Azure Machine Learning or a comparable platform, and MLflow or another model registry.
- Knowledge of reproducible Python environments, versioned data pipelines, evaluation automation, and online or batch deployment.
- Understanding of release approvals, monitoring, rollback, secure model-provider access, and embedding jobs.
- Ability to trace deployed models to evaluation evidence and distinguish infrastructure faults from model-quality concerns.
Desirable Experience
- Experience with Microsoft Foundry, Azure AI Search, RDF graph databases, and Temporal.
- Familiarity with Microsoft Fabric, data-lake services, and Purview for governed data workloads.
- Additional experience with GPU scheduling, self-hosted model serving, multi-region data residency, Microsoft Defender for Cloud, cost governance, and resilience testing.