Phd Researcher

Project background The increasing availability of multimodal health data, coupled with advances in AI, offers new opportunities to deliver personalised, scalable behavioural interventions in healthcare. However, current conversational AI systems remain limited in their ability to operate reliably across diverse clinical contexts. They often lack generalisability, struggle to integrate heterogeneous and longitudinal data, and do not adequately address key challenges related to safety, bias, robustness, and clinical validity. We are seeking a highly motivated PhD candidate to join an interdisciplinary Future Health Technologies (FHT) research programme at the forefront of agentic AI in healthcare. The project aims to design, develop, and evaluate AI-driven coaching and nudging systems capable of supporting positive patient outcomes across a range of conditions. The successful candidate will contribute to the development of a scalable and interoperable AI system that delivers personalised behavioural support across domains such as mental well-being, stroke rehabilitation, falls prevention, and chronic disease management. Central to this work is the integration of multimodal data ingestion, knowledge retrieval, and adaptive decision-making to deliver high-quality health/ behaviour change interventions with a strong emphasis on performance, safety, and real-world applicability. By bridging machine learning, behavioural science, and clinical research, the project seeks to establish foundational methods for trustworthy agentic AI systems that can be deployed across diverse healthcare settings. Job description The PhD will form part of a broader interdisciplinary research programme in mobile health and AI, with applications in areas such as mental well-being, fall prevention, stroke rehabilitation, and chronic disease prevention. The PhD student will work in an interdisciplinary team spanning behavioural scientists, clinical psychologists, computer scientists, AI engineers, and health systems researchers. The position involves close collaboration with research partners in Singapore, including Nanyang Technological University, National University of Singapore, Tan Tock Seng Hospital (TTSH) and Duke-NUS Medical School , as well as public health and healthcare stakeholders. The PhD student will be responsible for designing, developing, and validating a comprehensive agentic AI platform that supports multimodal data ingestion, knowledge retrieval, and LLM-based workflows for the delivery of evidence-based coaching strategies, including motivational interviewing. This includes analysing time-series data from smartphone sensors and usage logs using advanced statistical and machine learning approaches to develop personalised nudging and adaptive interventions. The candidate will also lead the development of a cross-platform mobile application for delivering and monitoring these interventions, including deployment and management of the associated backend infrastructure. The PhD student will work in close partnership with researchers developing hardware interfaces for device data ingestion and will support clinical validation studies to evaluate the feasibility, acceptability, and effectiveness of the developed interventions with end-users and key stakeholders. The candidate will also contribute to ethics submissions, data governance documentation, and analysis plans in accordance with Singaporean and international research standards. The role involves publishing findings in high-impact peer-reviewed journals and presenting at leading international conferences. Given the applied, translational nature of the project, the position offers exceptional opportunities for professional development and for contributing to healthcare innovations with direct real-world impact across domains including mental well-being, stroke rehabilitation, falls prevention, and chronic disease management. Selection Criteria You will have: Applicants must hold a Master’s degree (or be near completion) in Computer Science, Software Engineering, Informatics, Information Systems, or a closely related discipline. A strong theoretical and practical foundation in machine learning is essential, encompassing both the underlying principles, such as model evaluation and data pipelines, and their applied forms, including training, fine-tuning, and deployment of models using frameworks such as PyTorch or TensorFlow. Solid proficiency in Python and a genuine interest in…