CSIRO (Australia's National Science Agency) 
New South Wales, Australia Sintesi dell’opportunità
CSIRO offers a four-year Industry PhD Scholarship focused on the safe deployment and clinical monitoring of a multilingual AI triage system for linguistically diverse patients in Australian emergency departments. The project builds on the TRIBOT prototype, aiming to strengthen real-time inference pipelines and develop a safety-monitoring framework to detect failure modes before clinical translation. Located primarily at the University of New South Wales with industry engagement at Cogninet Australia, the role involves research and engineering to support simulation-based testing. The scholarship provides approximately $48,000 per annum tax-exempt. Ideal candidates possess strong foundations in computer science or software engineering, with demonstrated experience in machine learning, natural language processing, and Python frameworks like PyTorch or TensorFlow. Familiarity with large language models and healthcare data is highly regarded. Applications are open until the position is filled.
Generata dal contenuto memorizzato dell’opportunità. Verifica sempre i dettagli nell’annuncio ufficiale.
Descrizione del progetto
Funding and duration
four year scholarship package totalling approximately $48,000 per annum tax exempt (2026 rate); four-year scholarship and research training program.
Project overview
Project title
Safe Deployment and Clinical Monitoring of a Multilingual AI Triage System for Linguistically Diverse Patients in Emergency Departments
Project description
Australia’s emergency departments (“ED”) are under pressure. Each year, over eight million patients present to EDs nationally, including millions who speak little or no English. For these patients, the triage encounter, a critical 3 to 5-minute conversation that determines how urgently they receive care, is fraught with miscommunication, misclassification, and risk of harm.
TRIBOT is an AI-powered multilingual triage-assistance research prototype being developed at the University to support communication and structured information capture for linguistically diverse patients in ED triage. This Project will build on the TRIBOT program by focusing on the research, engineering and evaluation activities needed to strengthen the prototype’s real-time inference pipeline and clinical AI safety-monitoring framework.
The Project is intended to support simulation-based testing, approved research evaluation and future readiness for safe clinical translation. It does not, by itself, authorise live clinical deployment, clinical decision-making use, access to identifiable patient information, hospital implementation or integration with hospital systems. Any such activity will require separate ethics, governance, privacy, technical, regulatory, data access, work health and safety, and site approvals, as applicable.
The Student will research and engineer a prototype real-time inference pipeline capable of being tested against ED-relevant operational constraints, and design and evaluate a clinical AI safety-monitoring framework intended to detect uncertainty, failure modes, safety risks and escalation points before any clinical-use or deployment pathway is considered.
Supervisory team
University
Name of university supervisor | Padmanesan Narasimhan
Name of university | University of New South Wales
Email address | padmanesan@unsw.edu.au
Faculty | School of Population Health
CSIRO
Name of CSIRO supervisor | Cecile Paris
Email address | Cecile.Paris@csiro.au
CSIRO Research Unit | Technology
Industry
Name of industry supervisor | Prof Prabal Barua
Name of organisation | Cogninet Australia Pty Ltd
Email address | prabal@cogninet.com.au
Further details
Primary location of student | University of New South Wales, School of Population Health, 55 Botany Street, Randwick NSW 2031, Australia
Industry engagement component location | Cogninet Australia Pty Ltd, Level 5, 29-37 Bellevue Street, Surry Hills NSW 2010, Australia
Other locations | CSIRO Marsfield, 26 Pembroke Road, Marsfield NSW 2122, Australia
Ideal student skillset | A strong foundation in computer science, software engineering, or a related discipline, ideally with honours or masters-level research experience. Demonstrated experience in machine learning or natural language processing, with practical skills in Python and deep learning frameworks such as PyTorch or TensorFlow. Familiarity with large language models, inference optimisation, or MLOps pipelines will be highly regarded. Experience working with healthcare data or in clinical settings is desirable but not essential. Strong analytical and problem-solving skills are essential, as is the ability to communicate technical findings clearly to clinical and non-technical audiences. Be comfortable working in a fast-paced, multidisciplinary team spanning AI engineering, emergency medicine, and health policy. A genuine commitment to health equity and responsible AI development is highly valued.
Application close date | Open until position filled
Apply | Contact Padmanesan Narasimhan