Platform Lead

Who you are

• Extensive production experience deploying and supporting ML systems
• Proven track record of leading engineering teams
• Demonstrated experience with Generative AI and LLM deployment patterns
• A proven history of reducing cloud spend on large-scale AI clusters
• Experience with tools like MLflow, Kubeflow, LangSmith, or Phoenix
• Expertise in AWS/GCP/Azure cost tools, Kubecost, or Cloudability
• Extensive background of Kubernetes (K8s), Docker, and service meshes
• Expert knowledge of Terraform, Ansible, Jenkins, or GitHub Actions
• Proficient in Python, Bash, or Go
• Familiarity with Triton Inference Server, vLLM, or Hugging Face TGI

What the job involves

• We are seeking a Platform Lead to spearhead our ML Ops infrastructure, cost-optimisation, and deployment strategies. You will manage a talented team of DevOps engineers while remaining deeply technical and hands-on. Your primary mission is to build, scale, and secure the foundational platforms for our machine learning (ML) and generative AI (GenAI) models while maintaining financial accountability
• Team Leadership & Collaboration: Lead and manage a dedicated DevOps team, mentoring both junior and senior engineers while collaborating closely with Data Science and Engineering leaders
• Infrastructure Strategy & Automation: Define the infrastructure roadmap for AI/ML workloads and automate provisioning across cloud environments using Infrastructure as Code (IaC)
• MLOps & LLMOps Engineering: Architect, maintain, and optimise robust MLOps/LLMOps pipelines and CI/CD frameworks for continuous model deployment
• GenAI Production Deployment: Deploy Large Language Models (LLMs) into production environments, ensuring high availability, low latency, and optimal performance for GenAI applications
• FinOps & Budget Management: Establish FinOps frameworks to track, allocate, and forecast AI infrastructure spend, managing high-cost GPU/CPU cloud budgets
• Resource Efficiency & Unit Economics: Implement auto-scaling, spot instances, and down-scaling policies to eliminate waste, while providing full visibility into the unit economics of training and serving LLM models
• Observability & Incident Response: Establish 24/7 incident response, telemetry, and observability metrics to monitor system performance, model drift, and data pipelines
• Data Governance & Security: Enforce strict data governance, platform security, and compliance protocols across all AI/ML infrastructure