Miller School Unveils AI Platform to Transform Translational Research and Scientific Training
The University of Miami Miller School of Medicine has launched the MIL Agentic Data Scientist (MILADS), an AI platform built to help researchers plan better studies, analyze data faster and turn results into clinical and population health impact.
Developed by the Department of Informatics and Health Data Science, the Media and Innovation Lab and the Center for Translational Sleep and Circadian Sciences, the platform debuted at the 2025 PRIME Winter Bootcamp, an NIH-funded national training program for early-career investigators. Faculty leaders included Girardin Jean-Louis, Ph.D.; Tatjana Rundek, M.D.; and Debbie Chung, Ph.D., with Indu Ayappa, Ph.D., of the Icahn School of Medicine at Mount Sinai, on the PRIME leadership team.
An AI Partner for Scientists
MILADS is built as an intelligent collaborator. It supports investigators across experience levels-lowering the learning curve for newcomers and freeing up time for senior scientists by automating repetitive steps and orchestrating complex, multimodal analyses across clinical, behavioral, biological, environmental and real-world data.
The platform's agentic framework was led by Dwayne Henclewood, Ph.D., in partnership with Azizi Seixas, Ph.D., who also paired the technology with a structured training and mentorship model piloted at PRIME. At the bootcamp, more than 30 junior faculty and postdocs used MILADS on real research questions to improve study design, accelerate analysis and craft outputs ready for grant and manuscript development.
"The MIL Agentic Data Scientist platform fundamentally changes how we think about data science in academic medicine," said Dr. Seixas, associate professor of psychiatry and behavioral sciences; director of the Media and Innovation Lab; associate director of the Center for Translational Sleep and Circadian Sciences; and interim chair of the Department of Informatics and Health Data Science. "MILADS transforms biomedical research into a reproducible, agent-assisted system that serves as a teacher and thought partner, providing end-to-end support across the research pathway and accelerating ingenuity, grants and publications."
How the Platform Works
MILADS aligns to the full research lifecycle with five practical stages:
- Ideation: Shape questions, surface relevant datasets and stress-test early hypotheses.
- Intelligence: Synthesize multimodal evidence to see the full picture behind outcomes.
- Inception: Assist with study design, analytic selection and data preparation.
- Intervention: Run predictive modeling, classification, causal inference and time-series analysis to extract meaningful findings.
- Impact: Translate results into clinical tools, population health strategies, grant sections and manuscript-ready figures and text.
This structure moves projects from idea to implementation with clearer decisions, reproducible steps and transparent reasoning at each phase.
A Catalyst for Training, Mentoring and Workforce Development
The launch pairs technology with training. PRIME-led by Dr. Jean-Louis and nationally recognized for preparing early-career investigators in sleep, circadian biology and cardiometabolic health-has trained hundreds of scientists over two decades through mentorship, peer support and intensive methods education.
Embedding MILADS into PRIME modernizes that pipeline. Learners can:
- Practice real-world data analysis with guidance on best practices
- Build end-to-end, reproducible workflows they can reuse and share
- Strengthen grant and manuscript preparation with structured outputs
- Collaborate across disciplines using a common research framework
For context on federally supported researcher development, see NIH Research Training and Career Development programs at researchtraining.nih.gov.
What This Means for Your Lab
- Faster study planning with clearer analytic paths and fewer reworks
- Consistent data management and documentation for easier handoffs
- Access to advanced methods without rebuilding tooling from scratch
- Outputs formatted for decisions-clinical protocols, population strategies, grants and papers
A National Leader in AI-Enabled Science
The Miller School is expanding its footprint in AI-powered biomedical research-pairing MILADS with hands-on AI training, the PRIME Bootcamp and an Office of AI in Medical Education. Next steps include rolling the platform out across departments and new collaborations to speed discovery and scale workforce development in translational science.
Want structured upskilling for research roles? Explore curated AI course paths by job at Complete AI Training.
Your membership also unlocks:
