MLOps Engineer for AI Platform & LLMs

Anaplan Inc

Greater London

Hybrid

GBP 110,000 - 140,000

Full time

13 days ago

Get more replies from employers

Send a job-specific resume in minutes.

Job summary

Anaplan Inc is seeking a ML Ops Engineer to join our Platform Engineering team. You will design, scale, and maintain high-performance MLOps and LLMOps infrastructure supporting our AI-infused scenario planning platform.

You will collaborate with Data Scientists, ML Engineers, and Cloud Infrastructure teams to streamline model training, deployment, and inference while optimizing GPU utilisation, reliability, and cost-efficiency.

Qualifications

Hands-on production experience in DevOps, SRE, or Platform Engineering, with some experience dedicated to AI/ML infrastructure.
Proven track record of deploying, scaling, and operationalising machine learning models and LLMs in cloud-native production environments.
Demonstrated experience managing compute-intensive GPU infrastructure and HPC environments.
Advanced proficiency in Kubernetes (K8s), Docker, Helm, Kubeflow, and service meshes (Istio).
Hands-on experience with Terraform, Ansible, GitHub Actions, ArgoCD, or Jenkins.
Experience with vLLM, Ray, MLflow, LangChain / LangSmith, DeepSpeed, or Hugging Face TGI.
Solid background in AWS / GCP / Azure, Kubecost, and GPU cost optimisation techniques.
Strong skills in Python, Bash, or Go; deep knowledge of Linux kernel tuning and performance monitoring.

Responsibilities

Provision and manage cloud-native AI/ML infrastructure utilising Kubernetes, Docker, and GPU orchestration frameworks (e.g., NVIDIA GPU Operator, Slurm, or Ray).
Automate core platform infrastructure using IaC tools like Terraform, Helm, and Ansible.
Optimise GPU compute workloads, high-speed networking, and storage for efficient model training and low-latency inference.
Build and maintain robust CI/CD and MLOps pipelines for continuous model training, evaluation, packaging, and production deployment.
Deploy Large Language Models (LLMs) and generative AI workloads using advanced inference engines (e.g., Triton Inference Server, vLLM, TensorRT-LLM).
Enable automated model validation, monitoring for model drift, data drift, and latency bottlenecks.
Monitor and optimise cloud spend across high-cost GPU/CPU clusters across AWS, GCP, or Azure.
Implement auto-scaling strategies, spot instance policies, and dynamic resource allocation to eliminate infrastructure waste.
Establish benchmarking and telemetry to track unit economics and throughput for training and serving AI models.
Implement end-to-end observability using tools like Prometheus, Grafana, OpenTelemetry, and Weights & Biases or MLflow.