SnappyMeal
Context-aware, flexible nutritional logging using a multimodal AI application.
About
I'm a PhD student at the University of Cambridge working at the intersection of human-centered AI and healthcare. My research spans language models, multimodal sensing, and personalized health technologies, with a focus on making AI both effective and healthy to interact with.
Research interests
- Human-centered AI
- Designing intelligent systems that prioritize human agency and well-being.
- AI for health
- Turning patient-generated, physiological, and behavioral data into useful insights for people and clinicians.
- Multimodal intelligent systems
- Combining language models, sensing, and personalization to understand real-world health context.
Selected publications
-
Brenna Li, Liam Bakar, …, Alex Mariakakis, and Khai N. Truong. “Digitizing the Pre-consultation Experience: Impacts and Design Recommendations.” CHI ’26, 2026.
-
Liam Bakar, …, Shwetak Patel, and Vikram Iyer. “SnappyMeal: Design and Longitudinal Evaluation of a Multimodal AI Food Logging Application.” arXiv, 2025.
Selected projects
Multimodal cancer risk analysis
Exploring AI techniques and modalities for analyzing risk in cancer patients.
Pre-consultation chatbot
Addressing survey fatigue while giving physicians better context before consultations.
Writing
Game Theory IRL
Games are everywhere in our lives—but should we model our relationships around winning them?
Some pictures
Life through a lens—or several.
Are LLMs the new calculators?
Thoughts on language models, learning, and what classrooms should preserve.
USA Men’s Volleyball
The structural ceiling limiting men’s volleyball in the United States.
Contact
I’m open to conversations about research, projects, and collaboration.
lbakar@uw.edubakar.liam.20@gmail.com