Sr Director​/Scientific Fellow, AI Safety, R&D Data Science and Digital Health

At Johnson & Johnson, we believe health is everything. Our strength in healthcare innovation empowers us to build a world where complex diseases are prevented, treated, and cured, where treatments are smarter and less invasive, and solutions are personal. Through our expertise in Innovative Medicine and Med Tech, we are uniquely positioned to innovate across the full spectrum of healthcare solutions today to deliver the breakthroughs of tomorrow, and profoundly impact health for humanity.

Learn more at

As guided by Our Credo, Johnson & Johnson is responsible to our employees who work with us throughout the world. We provide an inclusive work environment where each person is considered as an individual. At Johnson & Johnson, we respect the diversity and dignity of our employees and recognize their merit.

Job Function: R&D Product Development

Job Sub Function: R&D Machine Learning

Job Category: People Leader

All Job Posting Locations: Beerse, Antwerp, Belgium, Zug, Switzerland

Job Description:

Johnson & Johnson Innovative Medicine (IM) is recruiting for a Scientific Fellow, AI Safety, R&D Data Science and Digital Health. This position can be located in New Brunswick, NJ;
Titusville, NJ;
Springhouse, PA;
La Jolla, CA;
Cambridge, MA;
Beerse, Belgium; or Zug, Switzerland. This position will require up to 25% travel.

Our expertise in Innovative Medicine is informed and inspired by patients, whose insights fuel our science-based advancements. Visionaries like you work on teams that save lives by developing the medicines of tomorrow.

Join us in developing treatments, finding cures, and pioneering the path from lab to life while championing patients every step of the way.

We are seeking a highly technical leader in AI safety for our Research & Development Data Science & Digital Health (DSDH) organization. Reporting directly to the Vice President of AI/ML & Digital Health, this role is responsible for embedding AI safety, robustness, and observability into the design, evaluation, and deployment of advanced AI systems across the DSDH portfolio and R&D use cases.

These systems span foundation and predictive AI models, generative AI, and autonomous agentic systems supporting discovery, development, clinical, and regulatory workflows.

This is a hands-on, technical, deeply scientific fellow role, focused on shaping model and AI system design and evaluation while contributing to policy, compliance, and enterprise governance. The Scientific Fellow will work closely with AI scientists, engineers, AI Quality & Optimization, Global Regulatory Affairs, Quantitative Scientists, and Johnson & Johnson Technology (JJT) to ensure AI systems deployed in R&D workflows are safe, trustworthy, and fit‑for‑purpose as AI capability and autonomy scale.

• Shape DSDH and IM R&D strategy for safe and trustworthy AI by defining multi-year research priorities, capability roadmaps, and investment recommendations for AI safety across discovery, development, clinical, and regulatory workflows.
• Represent AI safety as a senior scientific voice in function- and enterprise-level councils/working groups; set standards and priorities for safe scaling of GenAI and agentic systems, and provide technical leadership on safety principles and implementation for agentic and autonomous systems.

• Research, embed and implement AI safety‑by‑design principles into the development of foundation models, AI and generative AI applications, and agentic systems across R&D use cases.
• During all design phases, partner directly with AI and quantitative scientists across IM R&D, as well as with technical leads in JJT to:
  • Identify potential failure modes, risks, appropriate levels of autonomy and human oversight,
  • define safety‑relevant observability signals
    , acceptable failure envelopes and mitigation strategies tailored to different R&D contexts (research, clinical, regulatory)
  • ensure monitoring captures unsafe behaviors, not only performance drift.
• Design and execute safety‑focused models and evaluations
  , including but not limited to stress testing for hallucinations,…