Fullstack AI Quality Engineer – Applied & Agentic AI Systems

At Roche you can show up as yourself, embraced for the unique qualities you bring. Our culture encourages personal expression, open dialogue, and genuine connections,  where you are valued, accepted and respected for who you are, allowing you to thrive both personally and professionally. This is how we aim to prevent, stop and cure diseases and ensure everyone has access to healthcare today and for generations to come.

Join Roche, where every voice matters. The Position External Job Ad Text Job description We are seeking a highly skilled AI Quality Engineer specializing in testing Generative (and non-Generative) AI applications and agentic systems to join our team. The ideal candidate will have expertise in both traditional software testing and specialized testing methodologies for Large Language Models (LLMs)-based and AI-based systems.

As a fullstack Quality Engineer within a lean, agentic software team, you will cover the entire spectrum of quality assurance. You will span manual exploratory testing, traditional automated UI/API/MCP testing (across our Python and TypeScript stack), classical machine learning validation, and cutting-edge generative and agentic AI evaluation. This role requires a deep understanding of LLM, agentic and ML systems behaviors, and testing frameworks to ensure the reliability, fairness, and effectiveness of our AI-powered solutions.

Description of the area At Roche Digital Technology, we are advancing the boundaries of Applied AI. The Applied AI Engineering Team focuses on architecting, building, and operating high-value AI solutions and services to solve complex business challenges in healthcare. In the 2026 tech landscape, we operate in small, highly autonomous agile teams (e.g., ~9 members) powered by advanced coding agents (like Claude Code) to develop and ship solutions faster than ever before.

In this highly regulated environment, quality cannot be an afterthought. You will be the foundational pillar ensuring our rapidly developed agentic workflows and AI, GenAI, and agentic applications are safe, compliant, and robust before they reach the clinical or enterprise user. Job Responsibilities Generative & Agentic AI Testing Strategy: Design and implement comprehensive testing strategies specifically tailored for LLM-based applications and agentic systems, including evaluation of model responses, RAG pipelines accuracy, and overall system reliability.

You will expand this to evaluate autonomous multi-agent behaviors and tool-use accuracy. Full-Stack Test Automation: Design and implement automated testing solutions for continuous evaluation of LLM applications, including integration tests, performance tests, and specialized AI behavior tests. This spans both our Python backend services and modern TypeScript/React/Angular frontends. Quality Assurance Framework & Compliance: Utilize Roche's testing frameworks that address both traditional software quality aspects and AI-specific concerns such as output consistency, contextual accuracy, and ethical compliance; co-create and maintain such frameworks when required.

Create and maintain comprehensive test documentation testing strategy, test cases, and testing guidelines specific to AI applications and compliant with Roche practices. Edge Case, Bias, and Fairness Testing: Identify and develop test scenarios for edge cases in LLM behavior, including handling of ambiguous inputs, potential biases, and unexpected response patterns. Design and execute tests to identify potential biases in model outputs and ensure fair treatment across different user groups and use cases.

Security & Performance Testing: Collaborate within development teams to test for potential vulnerabilities specific to LLM applications, including prompt injection, data leakage, and other AI-specific security concerns. Collaborate with other engineers to conduct thorough performance testing of GenAI applications, including response time analysis, load testing, and resource