Platform Engineer

AI Platform Engineer - Senior Consultant - SC Eligibility required - Permanent

Locations: London, Manchester, Glasgow + other UK locations

Salary: Up to £70,000 + Bonus & Benefits

Active SC or SC Eligibility essential

As an AI Platform Engineer, you’ll design, build, and operate the infrastructure that enterprise AI and Generative AI workloads run on: the platform layer beneath LLMs, agents, and MLOps pipelines. This spans GPU-accelerated compute and container platforms, model serving and gateway infrastructure, evaluation and guardrail systems, and the MLOps/LLMOps tooling that takes a model from experiment to production. You’ll work across hybrid and multi-cloud environments, helping clients modernize their AI infrastructure and adopt AI safely and at scale.

As part of your role, you will:

• Be a senior or lead engineer on client AI platform engagements
• Architect and deploy AI-ready infrastructure (GPU-accelerated compute, Kubernetes/OpenShift, and cloud-native services) across cloud, on-premises, and hybrid environments
• Build and operate core AI platform components: model serving and gateway infrastructure, agent orchestration and tool-calling frameworks, evaluation harnesses, and guardrail/governance layers
• Implement MLOps and LLMOps pipelines (model deployment, monitoring, retraining, and fine-tuning where relevant) using Infrastructure-as-Code, GitOps, and CI/CD
• Establish observability, security, and governance frameworks specific to AI systems, including cost attribution and lifecycle management
• Work with clients and internal teams to develop new opportunities and shape a strong AI platform engineering culture
• Lead client workshops, architecture reviews, and technical briefings; provide operational support including monitoring and troubleshooting
• Share your knowledge and experience with colleagues as you coach and mentor them, while developing your own skills by experimenting with and learning new technologies

You’ll bring deep, hands-on experience in most of the areas below, with strong depth in AI/GenAI platform engineering specifically. You don’t need to tick every box.

• Model serving and gateway infrastructure (e.g. vLLM, LiteLLM, managed endpoints), with routing, failover, and per-workload cost attribution
• Agent orchestration and tool-calling frameworks (e.g. LangGraph or equivalent), including familiarity with the Model Context Protocol (MCP)
• Guardrail and AI-observability tooling (e.g. NeMo Guardrails, OpenTelemetry GenAI conventions, LangSmith, Braintrust)

MLOps & LLMOps

• Hands-on with MLOps platforms (Azure ML, Databricks, SageMaker) and vector/retrieval databases (Pinecone, Milvus, pgvector)
• Experience with GPU-accelerated infrastructure and NVIDIA AI Enterprise or equivalent stacks
• Exposure to fine-tuning, RLHF, or SLM distillation is a strong plus

Cloud-Native & Infrastructure

• Deep expertise in Kubernetes and container platforms (OpenShift, AKS, EKS, GKE, or VMware Tanzu)
• Infrastructure as Code and DevOps practices (Terraform, Bicep, Ansible, GitOps and CI/CD pipelines)
• 5+ years’ experience across Azure, AWS, or GCP; strong DevOps fundamentals