Data & validation manager
Pubblicato il 22-09-2026 - Cosmo I Building Health Confidence in Italia
At Cosmo, we are at the forefront of revolutionizing healthcare through our groundbreaking technology. With our innovative advancements in live endoscopy, we are transforming the landscape of medical diagnostics. We leverage AI to empower physicians, enabling them to make well-informed decisions and significantly improve the lives of their patients. Our dedication to excellence has resulted in creating GI Genius: our AI-enabled medical device for live endoscopy has received FDA approval and is successfully deployed in hospitals worldwide, making a significant impact in the field of healthcare.
Check out our website for more information at The Data & Validation Manager is responsible for coordinating and integrating data lifecycle and validation activities within the Med Tech AI Division, ensuring that high-quality, representative, and regulatory-ready evidence is available to support the development, validation, and release of AI-enabled medical technologies. The role provides stewardship across data governance, dataset readiness, and system validation,
working through specialist leads and senior engineers responsible for data collection, annotation, and test execution.
The Manager ensures consistency, traceability, and alignment of data and validation outputs with product needs, validation strategies, and regulatory expectations, without replacing domain-specific ownership of methods and execution.
Reporting to the R&D; Factory Director, the Data & Validation Manager contributes to the definition and coordination of validation strategies, statistical methodologies, and performance metrics for AI systems, and supports their coherent application across software-defined medical devices, platforms, and supporting tools.
The role acts as an integration point between Artificial Intelligence, Software Engineering, Hardware & NPI, Clinical Affairs, Service Engineering, Product Development Quality, and R&D; Operations, ensuring that validation evidence is reusable, comparable