Postdoctoral Fellow, TEAM-AI Lab, Department of Quantitative and Systems Health Sciences - Now Hiring

General Notes

Dell Medical School is seeking a Postdoctoral Fellow, TEAM-AM Lab for the Department of Quantitative and Systems Health Sciences.
 

Purpose

The TEAM-AI Lab seeks multiple Postdoctoral Research Associates to lead methodological innovation, software architecture engineering, and scientific execution across its active grant portfolio. Working under the direct mentorship of Dr. Hongfang Liu and lab faculty, the Postdoctoral Researcher will drive research at the intersection of health data science, multimodal AI, digital twins, computational phenotyping, and responsible AI. PhD must have been received within the last three years

The candidate will hold primary responsibility for designing novel algorithmic frameworks, coordinating multi-institutional research networks, and translating real-world health data into actionable clinical intelligence. This position provides structured preparation for an academic tenure-track career or lead research scientist position in industrial AI labs, providing access to national data networks, high-performance computing clusters, and clinical interdisciplinary collaborations across UT Austin.

The applicants will join a collaborative research environment at the Translational AI Excellence and Application in Medicine (TEAM-AI) Lab, focusing on accelerating the translation of AI innovations in biomedicine and healthcare. The lab consists of faculty members, program managers/coordinators, data scientists, and scientific programmers. The activities carried out by the team range from advancing AI innovations through big data, empowering biomedical and clinical sciences through team science collaboration and best practices, to building human-centered, value-added, and evidence-based tools, resources, and services to facilitate real-world implementation of said innovations. 

This position is a temporary with an end date of 08/31/27, renewable based upon availability of funding, work performance, and progress toward goals.
 

Grant reference

·       EMED: An Ethical Mixture-of-Experts Digital Twin Framework for Medical Device Surveillance  

·       CardioOnco-AI: AI-Empowered Cardiotoxicity Risk Prediction Among Breast Cancer Survivors Using Multi-Site Real-World Data:  

·       ReCARDO: Using Real-World Data to Derive Common Data Elements for Alzheimer's Disease and AD-Related Dementias Research Through Ontological Innovation:  

·       WONDER: Accelerating Real World Data-driven Precision Oncology through Data Science and Informatics Excellence in Research:  

·       POI-KB: Design and Development of a Knowledgebase for Accelerating Perioperative Organ Injury Research and Translation  

Responsibilities

·       Lead the design and implementation of mixture-of-experts neural architectures and reinforcement learning pipelines for counterfactual disease trajectory simulation for EMED, an NIH-funded multi-modal AI project.

·       Architect and evaluate multi-site cardiotoxicity risk prediction models integrating structured EHRs, clinical notes via natural language processing, strain echocardiography features, and non-medical determinants of health under the FDA CardioOnco-AI award.

·       Coordinate AI and computational phenotyping work streams within the national 10-institution ReCARDO network to extract, standardize, and validate Common Data Elements (CDEs) for Alzheimer's disease research.

·       Construct deep language models and clinical natural language processing pipelines to extract structured oncologic phenotypes, molecular biomarkers, and treatment responses from progress notes for the WONDER project.

·       Engineer semantic knowledge graphs and database query architectures capturing perioperative pathophysiological mechanisms for acute organ injury research under POI-KB.

·       Authorship of high-impact first-author or co-author manuscripts in leading informatics and machine learning journals and conferences.

·       Other related duties as assigned.

EDUCATION & EXPERIENCE

Minimum Qualifications:

• Ph.D. in Biomedical Informatics, Computer Science, Data Science, Electrical & Computer Engineering, Applied Mathematics, or a related quantitative field earned within the past three years.

• Strong background in one or more of the following areas:

  • generative and trajectory modeling including transformers, mixture-of-experts, reinforcement learning, simulation, and counterfactual analysis.  

  • Multimodal NLP & Fusion, large language models (LLMs), cross-attention Fusion, vision-language transformers.

  • Ontological engineering, knowledge graph construction and mining, CDE development for data harmonization. 

  • Regulatory science and explainable AI, verification, validation, uncertainty quantification, and AI evaluation framework

  • High-performance computing and big data analytics

• PhD must have been received within the last three years.

Preferred Qualifications:

• Demonstrated understanding of model validation, clinical trial design, and causal inference techniques.

• Experience with transformer-based models, LLMs, retrieval-augmented generation (RAG), or foundation models.

• Experience analyzing complex real-world clinical datasets (e.g., MIMIC-IV, OMOP CDM, PCORnet, NACC Uniform Data Set, or state cancer registries).

• Knowledge of causal inference, clinical prediction modeling, or multimodal AI.

• Experience with responsible AI, model evaluation, fairness, privacy, or explainable AI. 

• Experience with biomedical image processing, radiomics, or digital pathology.  

• Scientific programming on Linux or high-performance computing environments.

LICENSES, REGISTRATIONS OR CERTIFICATIONS

Required:

• None

Preferred:

• None

$63,480+ depending on NIH Level
 

WORKING ENVIRONMENT/EQUIPMENT

• Standard office equipment.

• Repetitive use of a keyboard.

• May be exposed to such occupational hazards as communicable diseases, blood borne pathogens, ionizing and non-ionizing radiation, hazardous medications and disoriented or combative patients, or others.

• May work in research laboratories, clinical environments, hospitals, ambulatory settings, or field research locations.

• May handle biological specimens, chemicals, hazardous materials, or laboratory equipment consistent with assigned research activities.

• May periodically lift and move research materials and equipment in accordance with organizational safety requirements.

• Requires visual acuity and manual dexterity sufficient to operate research equipment and computer systems.

• May require occasional evening, weekend, or travel commitments for research activities, conferences, and collaborative projects.

• Resume/CV

• 3 work references with their contact information; at least one reference should be from a supervisor

• Letter of interest

Important for applicants who are NOT current university employees or contingent workers: You will be prompted to submit your resume the first time you apply, then you will be provided an option to upload a new Resume for subsequent applications. Any additional Required Materials (letter of interest, references, etc.) will be uploaded in the Application Questions section; you will be able to multi-select additional files. Before submitting your online job application, ensure that ALL Required Materials have been uploaded.  Once your job application has been submitted, you cannot make changes.

Important for Current university employees and contingent workers: As a current university employee or contingent worker, you MUST apply within Workday by searching for Find UT Jobs. If you are a current University employee, log-in to Workday, navigate to your Worker Profile, click the Career link in the left hand navigation menu and then update the sections in your Professional Profile before you apply. This information will be pulled in to your application. The application is one page and you will be prompted to upload your resume. In addition, you must respond to the application questions presented to upload any additional Required Materials (letter of interest, references, etc.) that were noted above.