Mark your calendars for September 23rd! CTML Faculty member Alejandro Schuler will lead a workshop on "Agentic AI for Biostatistics Research" as part of CTML's Fall Seminar Series! This workshop will take place on Wednesday, September 23 at 12:00 PM in Berkeley Way West, 5th Floor, Room 5401.
Abstract: Agentic tools like claude code or codex are increasingly used in research. In this workshop, I will first explain what these tools are and how to set them up and use them. I will then discuss best practice for biostatistics research. On the logistical side, I'll discuss how to set up project context, use and build skills and agents.md files, and how to manage code, paper drafts, and results. I'll also discuss what kinds of research tasks agents are useful for today and how to make the best use of them. In particular, I will explore literature review, setting up and running simulation studies, proof-writing, and interpretation of results. Although modern LLMs are very capable executors, they still often struggle to see the big picture, distinguish between important and trivial findings, write coherently, or design specific, targeted experiments. This creates a risk of short-circuiting your research. Nonetheless, using these tools to rapidly execute experiments and proof can help you build towards meaningful results if you are patient and focused on your own judgement and learning.
Bio: Alejandro Schuler is an Assistant Professor in Residence at UC Berkeley Biostatistics. His research focus is on developing methods for clinical decision-making in the real world that are economically or clinically necessary, statistically rigorous, and frictionless from the user perspective. He completed his Ph.D. at Stanford in 2018 and worked as a postdoc with CTML before starting on the faculty. Dr. Schuler is known for developing the Selectively Adaptive Lasso, NGBoost, and prognostic covariate adjustment methods, among others. In addition, he often collaborates with domain experts to translate their questions to mathematical formalisms and bring the right methods to bear on them. His experiences working as a data scientist at Kaiser Permanente's Division of Research and as an early employee of a health tech startup helped shape his research agenda into something with relevance beyond academia. Dr. Schuler is also passionate about pedagogy and making good statistics accessible to everyone regardless of background or experience.