Dessi Zaharieva, PhD, CEP, CDCES
Assistant Professor
Stanford University School of Medicine
Featured in the Session: Getting Along with Your Algorithm
When
Sunday, June 7
at 4:30 p.m.
Where
Hall E-2 (Level 1)
Ernest N. Morial Convention Center

What is your presentation about?
This presentation will focus on how people with type 1 diabetes can work with their automated insulin delivery (AID) systems—rather than against them—especially in the context of exercise. The session will break down how these algorithms function in simple, practical terms and highlight why exercise remains one of the biggest challenges for AID performance. This session will include real-world, evidence-informed strategies for adjusting AID settings, anticipating glucose responses, and using key features safely and effectively. The goal is to translate complex technology into actionable approaches that reduce burden and improve confidence during exercise.
How do you hope your presentation will impact diabetes research or care?
I hope this presentation will help to bridge the gap between rapidly advancing diabetes technology and real-world use, particularly in the context of exercise where the guidance is sometimes lacking and still evolving. By simplifying how AID systems behave and offering practical strategies, the session aims to improve user confidence and safety, and simplify day-to-day decision-making around exercise.
How did you become involved with this area of diabetes research or care?
My background is in exercise physiology, and my early research focused on preventing hypoglycemia during exercise in people with type 1 diabetes. Over time, my work evolved alongside advances in continuous glucose monitoring (CGM) and AID systems, with a focus on making these technologies more effective during exercise. Through both research and structured exercise education, I’ve become particularly interested in translating complex guidelines into practical strategies that reduce the burden and help individuals feel more confident when being active.

