Director, AI/ML
AstraZeneca3 days ago
Bengaluru, IndiaStaff+
Responsibilities
- Define the technical vision, roadmap, architecture, and standards for traditional ML, generative AI, and agentic AI capabilities.
- Provide hands-on technical leadership, set design direction, review architectures and code, and resolve complex technical problems.
- Design and deliver multi-agent systems and LLM applications involving orchestration, tool use, memory, guardrails, and human-in-the-loop patterns.
- Establish automated and human-in-the-loop evaluation, benchmarking, and pre- and post-release quality gates.
- Drive fine-tuning, distillation, reinforcement learning, and preference-optimization approaches to improve model quality, cost, and latency.
- Set standards for telemetry, observability, monitoring, drift detection, and production cost management.
- Evaluate emerging models and frameworks, conduct proofs of concept, and translate them into production-grade solutions.
- Partner with platform, data engineering, architecture, security, and responsible-AI teams on governance, compliance, and reuse.
- Communicate technical concepts and trade-offs to senior non-technical partners and shape AI investment use cases.
- Mentor senior data scientists and ML engineers and represent AI work in internal and partner engagements.
Requirements
- Bachelor's degree or equivalent experience in computer science, mathematics, statistics, engineering, physics, computational sciences, or a related quantitative field.
- Typically 12+ years of experience building and deploying data science and machine learning solutions.
- Typically 2+ years of recent hands-on experience with production LLM-based and agentic applications.
- Demonstrated ability to set technical direction and lead complex, multifunctional AI delivery through influence rather than headcount.
- Expertise in prompt and context engineering, retrieval-augmented generation, fine-tuning, distillation, model selection, and cost and latency optimization.
- Experience designing multi-agent systems using Amazon Bedrock AgentCore or equivalent frameworks such as LangGraph, Strands, or AutoGen.
- Knowledge of LLM and agent evaluation, telemetry, tracing, and AI-system monitoring methods and platforms.
- Strong classical machine learning, statistical modeling, deep learning, and transformer experience using PyTorch or TensorFlow.
- Expert Python programming skills plus SQL, scikit-learn, Jupyter, and Streamlit or similar rapid-prototyping tools.
- Experience with AWS, SageMaker, Bedrock, MLflow or equivalent tracking tools, and Docker-based containerization.
- Preferred qualifications include a master's degree or PhD, reinforcement learning experience including RLHF or RLAIF, knowledge graphs, NLP, structured and unstructured data, Airflow, Spark, data-quality frameworks, relevant pharmaceutical or healthcare analytics experience, and AI/ML publications, patents, or open-source contributions.
- Strong technical judgment, communication, collaboration, curiosity, and ability to balance innovation with pragmatism, risk, and cost.
Benefits
- Average minimum of three days per week working from the Bangalore office, with flexibility around individual circumstances.
- Opportunity to work at enterprise scale on AI solutions supporting life-changing therapies and patient outcomes.
- Collaboration with internal specialists and external partners using advanced tooling and opportunities to move prototypes into production.
- Support for professional growth, technical craft, and profile development in an inclusive and diverse environment.
Categories
About AstraZeneca
AstraZeneca is a global biopharmaceutical company that discovers, develops, manufactures, and commercializes prescription medicines for patients and healthcare systems. Its portfolio spans oncology, cardiovascular, renal and metabolism, respiratory and immunology, and rare diseases (expanded by the 2021 acquisition of Alexion). Founded in 1999 and headquartered in Cambridge, UK, it is a public company listed on the London Stock Exchange and Nasdaq, generating revenue primarily from drug sales and partnerships.