Associate Director - AI ML and Cloud Security - AI Platform Engineering

Overview

Build and own the AI and ML platform infrastructure for Cyber Architecture and Solutions, and lead engineering delivery of the data fabric pipeline, translating AI Architecture's data fabric design into built, tested and vendor-ready infrastructure.

What you'll do

• Own the AKS and GKE clusters, GPU node pools and model-serving tier that every Cyber Architecture and Solutions agent runs on, including capacity, upgrade and resilience planning.
• Lead engineering delivery of the data fabric pipeline, translating architecture specifications into built, tested, vendor-ready infrastructure.
• Own Kubernetes and container orchestration for AI and ML workloads.
• Own CI/CD pipelines for AI and ML delivery.
• Own data ingestion and transformation layers, and govern data ingestion into the fabric.
• Integrate security controls into the platform by design, to the binding AI security standards.
• Prepare data fabric and platform handover packages so the vendor can operate them from day one.
• Own the contract between the platform and agent layers, including model-serving endpoints, vector index service levels and the token and cost budget per agent, and renegotiate it when an agent outgrows it.
• Lead the platform engineers delivering the data fabric and model-serving tier, and keep enough of your own week on the estate to make credible design calls.
• Run the shadow and reverse-shadow period where the vendor operates the data fabric under your supervision, and hold the handover gate until their runbook survives an injected failure.
• Lead your team in using approved AI tools to improve productivity and ways of working, role-modelling responsible use and following GSK's Responsible AI, privacy and security requirements.
• Own cloud security posture across Azure and GCP for the AI estate, including posture findings, misconfiguration remediation routing and the standards the platform is held to.
• Build and maintain the platform through infrastructure as code, so every environment is reproducible and every change is traceable.

What you'll need

• 15+ years in platform, cloud or AI security engineering in a regulated environment.
• Leading AI/ML platform engineering: Kubernetes and CI/CD for AI/ML workloads, GPU and compute provisioning, model serving, vector database operations, MLOps/LLMOps pipelines, model and prompt versioning and observability.
• Building the data fabric pipeline to specification with security by design: encryption at rest and in transit, RBAC, audit logging and classification enforcement, and owning data ingestion governance and change control.
• Managing the vendor's agentic transformation of the data fabric and vendor readiness for operations handover, ensuring GSK requirements, classification controls and resilience standards are maintained.
• GPU scheduling and model-serving operations at scale: KServe or vLLM behind an autoscaled Kubernetes pool, with p95 latency and cost per thousand tokens held to a published budget.
• Bachelor's degree in a relevant discipline, or equivalent professional experience.

Nice to have

• Terraform, infrastructure as code and cloud-native tooling, with vector databases and AI platform components.
• Cloud IAM, workload protection and AI workload detection controls.
• Compute and data pipeline infrastructure supporting autonomous AI agent operation.
• GPU and compute cost management and capacity planning for AI workloads.
• Running a large-scale ingestion or data fabric pipeline in production, including backfill, schema evolution and the cost curve as volume grows.

Details

• Location: Bengaluru Luxor North Tower.