Data Scientist

Roche Digital Technology (RDT) is a global team at the heart of Roche, focused on shaping the digital future of healthcare using artificial intelligence, data, and scalable tech innovations. The role is based in Roche's AI Function, building scalable platforms, tools, and GenAI capabilities for healthcare products impacting millions of patients worldwide.

• Strong hands-on proficiency in Python, LLMs, RAG pipelines, vector databases/embeddings, and NLP/statistical modeling with clean, reproducible code (Required).
• Proven experience building evaluation harnesses for GenAI, model fine-tuning, and implementing LLMOps/MLOps pipelines with full observability (Required).
• Skilled in communicating technical AI concepts to non-technical stakeholders and managing infrastructure and cost optimization (Required).
• Hands-on experience deploying scalable enterprise AI solutions on AWS or Azure platforms (Preferred).
• Experience with AI governance, model risk management, fairness/bias mitigation, or regulated domains like healthcare, pharma (GxP), or financial environments (Preferred).

• Design, build, and deploy production-ready AI/ML solutions and agentic workflows.
• Transform complex business problems into effective modeling approaches using classical ML, fine-tuned models, RAG, or agentic architectures.
• Establish evaluation frameworks, golden datasets, and testing harnesses for GenAI/LLM outputs.
• Build MLOps/LLMOps pipelines for continuous integration, deployment, and real-time observability.
• Embed AI governance, security, and compliance guardrails into development pipelines.
• Collaborate with product owners, engineers, and business leaders to communicate AI trade-offs and insights.

The role requires presence in Roche offices on average two days per week as part of a hybrid work model. The controller of personal data is Roche Polska Sp. z o.o. The company complies with law and has a procedure for reporting violations. Roche is an Equal Opportunity Employer.