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 … 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 ...