by Philip J. Freda and Attri Ghosh
11/18 (Wed) 9:00 - 13:00 (Kwak Joung-Hwan Hall, B146)
A hands-on tutorial series for computational biologists who want to build, customize, and orchestrate AI agents. Attendees will learn theory, skills, and tools to develop production-ready multi-agent pipelines. All examples are grounded in real biomedical tasks: data engineering, hypothesis generation, data analysis, and results interpretation.
By the end of this tutorial, you will be able to:
A month before the workshop, we will provide a GitHub link with all
requirements and package dependencies. However, these items will be required to
participate in the live demo:
Attri Ghosh, MS
Attri is a Research Data Scientist at Cedars-Sinai Health Sciences University in Los Angeles, USA. Her research focuses on translational biomedical informatics, with an emphasis on developing artificial intelligence and machine learning methods to accelerate biomedical discovery and advance precision health. Her work integrates multimodal biomedical data, including electronic health records and genomic data, to generate clinically meaningful insights, improve disease risk prediction, and support evidence-based clinical decision-making. Her overarching goal is to develop trustworthy, interpretable, and scalable AI systems that bridge computational innovation and real-world clinical applications.
Email: attri.ghosh@cshs.org
GitHub: https://github.com/AttriGhosh96
Google Scholar: https://scholar.google.com/citations?hl=en&user=4EF_nqwAAAAJ
Philip J Freda, PhD, MS
Phil is a clinical informatician, computational biologist, and AI researcher passionate about making healthcare data work harder for patients. His research focuses on building intelligent systems that can navigate the complexity of real-world clinical data, from messy EHR records to unstructured clinical notes, and transform them into actionable insights for clinicians and researchers. He is currently focused on building agentic AI systems that serve as collaborative partners in the research process itself, coordinating multi-step analytical workflows, automating data cleaning and engineering, and lowering the barrier between clinical questions and the computational methods that answer them. His work also extends to computational genetics, where he develops evolutionary computation approaches for uncovering complex genetic associations and non-additive genetic variation.
Email: philip.freda@csmc.edu
Website: http://philipfreda.com/
GitHub: https://github.com/PFreda-Lab
Google Scholar: https://scholar.google.com/citations?user=1NaI6RgAAAAJ