Lead Platform DevOps Engineer

Working Environment

You'll operate in production-focused environments where platform usability and operational excellence matter. The work sits at the intersection of platform engineering, DevOps and MLOps, with close collaboration across data science, ML engineering, software engineering and architecture.

The role requires comfort working 'in the weeds' - understanding how platforms behave under real workloads, and designing guardrails that balance flexibility for users with reliability, security and governance.

What You'll Be Doing

• Provide technical leadership across platform engineering, DevOps, and MLOps activities
• Design, build, and operate a Kubernetes-based MLOps platform that supports the full model lifecycle
• Implement and operate MLOps tooling and frameworks enabling teams to build, train, deploy, and serve models
• Develop and support model serving and inference capabilities within Kubernetes environments
• Implement workflows that support model experimentation (including notebooks), packaging, deployment and versioning
• Enable scalable inference and LLM-based workloads, including serving and optimisation considerations
• Work with data scientists and ML engineers to ensure the platform is usable, well documented and fit for purpose
• Own platform operability, reliability, security and supportability in live production environments
• Troubleshoot complex platform, workload and deployment issues across Kubernetes and MLOps layers
• Contribute to architectural decisions while remaining deeply hands-on in delivery

Your Experience

To be successful in this role, you will bring:

• Strong experience as a Senior or Lead Platform Engineer / DevOps Engineer
• Deep hands‑on experience building and operating Kubernetes-based platforms
• Strong practical experience using Helm and infrastructure-as-code tools such as Terraform
• Proven experience extending Kubernetes with higher-level platforms and services, rather than treating it as an end in itself
• Strong understanding of operational concerns: monitoring, logging, incident response, reliability and maintenance
• Confidence working directly with engineers and data scientists to support real workloads in production

MLOps experience is highly desirable, including exposure to tools and patterns such as:

• Building MLOps platforms using frameworks such as Kubeflow (or comparable approaches)
• Operating model serving and inference platforms (e.g. KServe, vLLM, or comparable solutions)
• Supporting LLM-based workloads, including optimisation and serving considerations
• Providing notebook-based development environments (e.g. JupyterHub) within secure platforms
• Exposure to emerging tooling such as InstructLab, trustworthy AI tooling, or equivalent approaches

In Return

You'll play a key role in enabling AI delivery at scale by building platforms that other engineers and data scientists actually want to use.

This is an opportunity to lead technically, shape a practical MLOps platform, and own operational outcomes in production environments where reliability and usability matter.