Senior AI Engineer

Senior AI Engineer

Forward Deployed AI, Enterprise AI Unit. Cambridge, UK

Competitive salary and benefits

Are you ready to turn ambiguous enterprise challenges into credible, production AI that moves the needle for patients and the business? Do you want to ship agentic and retrieval-augmented systems that colleagues rely on every day? Join our forward‑deployed AI engineering team within a newly created enterprise unit in Cambridge, where your work will help accelerate how we discover, develop and deliver medicines.

Accountabilities

• Agentic AI Development: Design, implement and iterate agentic workflows that orchestrate tools, sub‑agents and multi‑step reasoning to deliver deterministic, auditable outcomes to business users.
• Retrieval‑Augmented Systems: Build robust RAG pipelines and context engineering strategies that ground LLM outputs in enterprise knowledge, improve factuality and reduce hallucinations.
• Evaluation and Quality: Establish ground‑truth datasets, LLM‑as‑judge and rule‑based evals, and regression suites that separate signal from prompt noise and ensure consistent performance over time.
• Production Software Engineering: Translate prototypes into reliable services with structured outputs, clear interfaces, packaging, testing and traceability that meet critical software standards.
• Platform Integration and Observability: Integrate with Kubernetes and cloud platforms, implement CI/CD, and instrument systems for cost, latency, caching and evaluation dashboards to sustain performance at scale.
• Partner Teamwork: Work with subject‑matter experts and product owners to turn ambiguous problems into clear specifications, communicate trade‑offs, and land solutions that create business value.
• Secure and Scalable Deployment: Apply gateway‑based inference patterns and led LLM platforms where appropriate to balance security, performance and maintainability.
• Continuous Innovation: Scan the evolving AI landscape, assess when fine‑tuning or distillation is warranted, and proactively upgrade techniques to keep solutions robust and competitive.

Essential Skills/Experience

• Advanced university degree in mathematics, computer science or another relevant numerical/computational subject area; alternatively, equivalent experience in relevant research or industry. PhD or equivalent experience desirable.
• Deep and validated AI foundations — ability to go deep into the maths and nuances of neural networks, transformers, statistics, evaluation methodology — applied with judgement.
• Hands‑on experience designing and shipping production‑grade systems built around LLMs: agentic workflows, RAG, tool use, structured outputs and multi‑step orchestration.
• Practical experience with modern agent‑integration patterns — MCP, tool/skill ecosystems, agent‑to‑agent communication, sub‑agents and context engineering — and judgement about when to use each.
• Demonstrated ability to deliver production‑grade software using AI coding assistants with the subject area that critical software demands — specs, tests, evals, code review, traceability — not just ad‑hoc prototyping.
• Excellent evaluation skills: designing ground‑truth datasets, building LLM‑as‑judge and rule‑based evals, regression suites, calibration and statistical reasoning to distinguish signal from prompt noise.
• Strong Python — structured‑output validation, packaging — and focused testing practice; confident scaffolding a service from prototype to deployed.
• Excellent problem‑solving, collaboration and communication skills; able to translate ambiguous business problems into specs, collaborate with SMEs, and present results to senior stakeholders.
• Experience with Kubernetes and cloud platforms (Azure and/or AWS), CI/CD, and LLM observability (cost, latency, caching, eval dashboards).

Desirable Skills/Experience

• Experience training, fine‑tuning or distilling models when the problem genuinely calls for it.
• Familiarity with managed‑LLM platforms (e.g. AWS Bedrock, Azure OpenAI) and gateway‑based inference patterns. End‑to‑end software life cycle and experience shipping and deploying production code.
• Full‑stack delivery experience — comfortable leading a thin frontend when needed; this is not a frontend role and professional UI skills are not required.
• Interest in working on real‑world enterprise problems in a research‑driven organisation; pharma or life‑sciences experience welcomed but not required.

Our mission is to build an inclusive and equitable environment. We want people to feel they belong at AstraZeneca and Alexion, starting with our recruitment process. We welcome and consider applications from all qualified candidates, regardless of characteristics. We offer reasonable adjustments/accommodations to help all candidates perform at their best. If you have a need for any adjustments/accommodations, please complete the section in the application form.

#J-18808-Ljbffr…