Associate Director, RDU IT - MLOps

About the role

AstraZeneca is looking for an Associate Director, RDU IT - MLOps, based in Barcelona with an on-site commitment of three days a week and fluency in English. Reporting to the AI Director, you will lead the strategy, architecture, delivery and governance of the enterprise AI platform and accelerators, defining reusable services, setting engineering standards and embedding responsible AI so teams can deploy agentic and generative AI safely and cost-effectively.

Key responsibilities

• Own the product vision, architecture, roadmap and delivery of the RDU AI/ML platform, including reusable frameworks, reference architectures, APIs, templates and self-service services
• Ensure the platform is secure, resilient and scalable, using infrastructure as code, monitoring, incident management and disaster recovery
• Define and govern engineering standards, secure CI/CD, automated testing, code quality, vulnerability management and observability
• Lead solution design and trade-offs across security, integration, data, performance, scalability, interoperability and cost
• Own MLOps and LLMOps pipelines, automating model and LLM workflows and integrating them with enterprise data and applications
• Own the cost model, tracking cost per inference, GPU utilisation and token economics, and run showback/chargeback to internal consumers
• Embed security, privacy, data governance, model risk management and responsible AI controls, in line with GxP and regulatory standards
• Drive the adoption of AI in IT operations (observability, AIOps, AgentOps, service management)

Required profile

• Bachelor's degree in Computer Science, Electrical Engineering, Mathematics, Statistics or a related field
• 8+ years in software, cloud, DevOps or AI/ML engineering, with 4+ years deploying and maintaining ML models in production
• Proven technical leadership experience delivering platform capabilities
• Desirable: experience in the pharmaceutical industry, a Master's degree or ongoing advanced study

Required skills

• Hands-on experience with serverless architectures, Kubernetes, Docker and Terraform/Terragrunt
• CI/CD tools such as GitHub Actions, Jenkins or Bitbucket pipelines
• Orchestration tools such as Airflow, Flyte or Kubeflow
• Vector databases, RAG patterns, LangChain/LlamaIndex, and MLOps tools such as MLFlow, Sagemaker or Azure
• Python with TensorFlow, Keras or PyTorch
• Cloud platforms such as AWS, Azure or GCP