AstraZeneca Healthcare/BioTech

Associate Director (w/f/d), Agentic AI Platform, AI for Science Innovation (AISI)

Munich, Germany In person Full-time

Location: AZ Computational Pathology GmbH - Munich, Germany

About AstraZeneca

AstraZeneca is a global, science-driven biopharmaceutical company. We are dedicated to discovering, developing, and delivering innovative, meaningful medicines and healthcare solutions that enrich the lives of patients.

Site Description - Munich, Germany

Welcome to our Munich site, providing a collaborative environment where everyone can feel comfortable being themselves, a value at the core of AstraZeneca's culture. To help you maintain your best self, we offer after-work events, lunch & learns, a spacious and sustainable office environment, family and childcare support, with the Alps just around the corner for hiking, biking and skiing.

We are looking for an experienced

to join our Enterprise AI organization.

AI for Science Innovation (AISI), part of Enterprise AI, builds and runs AstraZeneca's agentic AI platform for research and development. The platform is in production with thousands of users across the company, who use it every day through a web application, and it connects to enterprise data sources and scientific tools through MCP servers and toolkits.

Main Duties and Responsibilities

You will join the platform engineering team that owns the platform's core. In this role, you will:

Build and operate the core of a production agentic AI platform used across the company.

Build MCP servers, A2A agents and tools that connect agents to enterprise systems and scientific data platforms.

Develop agentic AI applications for enterprise and scientific use cases, and measure their quality with evaluations.

Implement guardrails and data protection in the platform.

Run the platform in production: releases, reliability, performance, cost and observability.

Essential Requirements

Graduate degree in Computer Science, Engineering, or a related quantitative field.

5+ years of software engineering experience, including building and operating production systems.

Hands-on experience building and running an enterprise agentic AI platform in production with thousands of users, including the agent runtime, conversation and run management, tool calling, and operation at scale (concurrency, cost, reliability). Building a single agent or a prototype for a small group of users does not meet this requirement.

Experience building agentic AI applications for both enterprise and scientific use cases.

Experience building MCP servers or equivalent tool integrations for LLM agents against enterprise APIs.

Experience with harness engineering for LLM agents: system prompts, skill loading, progressive disclosure of tools, and context compression.

Experience designing and running evaluations of LLM agents (tool-usage evaluators, LLM-as-judge, eval datasets).

Experience implementing guardrails and data protection for LLM agents: access control, credential and secret handling, egress controls, data source classification, private-data handling, and file-system isolation of user workspaces.

Strong Python, including asynchronous programming (asyncio) and Pydantic.

Experience with TypeScript and a modern front-end framework (React).

Experience deploying and operating services on Kubernetes, with CI/CD and unit, integration and end-to-end tests.

Experience integrating LLM gateways, including API key management.

Experience with PostgreSQL (including connection pool tuning) and REST / WebSocket APIs.

Experience with observability for LLM systems: OpenTelemetry tracing, dashboards for errors, tool calls and token consumption (New Relic or equivalent).

Desirable

  • Experience with Pydantic-AI, AG-UI or similar agent frameworks.
  • Experience with the A2A protocol and multi-agent systems.
  • Experience with multiple model providers (AWS Bedrock, Anthropic, Google).
  • Experience with data protection controls for AI systems in a regulated industry.
  • Experience with Knative or serverless deployment on Kubernetes.
  • Experience in a pharmaceutical or life sciences R&D environment.

What you can expect

  • The opportunity to shape and scale a production agentic AI platform used across AstraZeneca.
  • Work at the intersection of AI, software engineering and scientific R&D.
  • Collaboration with AI engineers, scientists, technology teams and cross-functional partners across the organization.
  • The opportunity to solve complex engineering challenges around agentic AI, enterprise integration, reliability, evaluation and data protection.
  • An environment that encourages technical excellence, knowledge sharing and continuous development.

Date Posted

17-Nov.-2026

Closing Date

AstraZeneca embraces diversity and equality of opportunity. We are committed to building an inclusive and diverse team representing all backgrounds, with as wide a range of perspectives as possible, and harnessing industry-leading skills. We believe that the more inclusive we are, the better our work will be. We welcome and consider applications to join our team from all qualified candidates, regardless of their characteristics. We comply with all applicable laws and regulations on non-discrimination in employment (and recruitment), as well as work authorization and employment eligibility verification requirements.

Job description provided by AstraZeneca.
View the original posting