Postdoctoral Researcher Video team

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

Job Function:

Career Programs

Job Sub Function:

Post Doc – Data Analytics & Computational Sciences

Job Category:

Career Program

All Job Posting Locations:

Zug, Switzerland

Janssen Research & Development LLC, a Johnson & Johnson company, is recruiting a postdoctoral researcher to join the Video Understanding team. Positions are available in the US (Titusville, NJ; Raritan, NJ; La Jolla, CA; Cambridge, MA; New York, NY; Spring House, PA) or Europe (UK, Netherlands, Switzerland, Austria). Remote arrangements will also be considered.

Janssen develops treatments that improve the health of people worldwide. Our research spans oncology, cardiovascular and metabolic disorders, immunology, neuroscience, and infectious disease. Our goal is to help people live longer, healthier lives. We have produced and marketed many first-in-class prescription medications and are poised to serve the broad needs of the healthcare market – from patients to practitioners and from clinics to hospitals.

We are seeking highly skilled and motivated Postdoctoral Researchers to join our Video Understanding team at Johnson & Johnson. In this role, you will design, implement, and evaluate state-of-the-art AI/ML methods for frame-level and video-level understanding across diverse medical modalities, including endoscopy, ultrasound, and magnetic resonance enterography (MRE). Your work will specifically focus on developing robust uncertainty models and multi-modal architectures to extract reliable, high-confidence insights from clinical data.

You will collaborate closely with clinicians, engineers, and data scientists to translate these advanced algorithms into robust, interpretable, high-impact tools that optimize decision-making within clinical trial efficacy and workflows.

• Conceive, develop, and implement ideas with key internal clinical personnel to understand needs and use cases around AI tools for endoscopy.
• Design, develop, and evaluate deep learning models for medical video and image understanding (scoring, segmentation, and detection).
• Lead the research and development of uncertainty quantification methods for model predictions (e.g., Bayesian approaches, ensembles, calibration metrics, and predictive intervals) and integrate uncertainty estimates to quantify the reliability and confidence of AI-driven diagnostic model outputs.
• Develop advanced multi-modal models for video analysis, integrating data from endoscopy, ultrasound, intestinal ultrasound (IUS), and Magnetic Resonance Enterography (MRE).
• Clearly articulate highly technical methods and results to diverse audiences and partners to drive decision-making.
• Participate in cross-functional team meetings, drive discussion and follow-up questions to collaborators, and compile answers into briefing reports.
• Extract insights from collection of briefing reports focusing on business value of AI pipelines for immunology.

Required Qualifications:

• A Ph.D. degree in a quantitative discipline (e.g., Physics, mathematics, computer science, electrical engineering, or similar).
• Demonstrated experience driving research in and applying Computer Vision techniques (e.g., Transformers, CNNs, RNNs, GANs).
• Proven expertise in uncertainty model and measures development, including techniques for uncertainty quantification in deep learning.
• Demonstrated experience on state-of-the-art techniques for video understanding (e.g., foundational models, transformers).
• Proficiency with one or more programming language such as Python or C++.
• Extensive experience with traditional Computer Vision applications, such as OpenCV, object detection, edge detection, image…