# Ernest Lim
Hi, I'm Ernest - a clinician scientist based in London 🇬🇧, focused on making conversational AI safe and effective for healthcare.
I'm currently seconded from my surgical training in ophthalmology to serve as Science Director at Ufonia, where I lead research on autonomous clinical conversations using generative AI.
I'm also completing a PhD in Computer Science at the University of York's Centre for Assuring Autonomy, exploring how to apply safety engineering to real-world AI systems in healthcare.
Outside of research, I'm passionate about helping others build with AI. I teach doctors to code through my course clinician2creator, and co-founded stinterview to mentor aspiring ophthalmologists. I occasionally write about productivity, health-tech, and learning. When I'm not working (or vibe-coding), you'll probably find me cooking or on a tennis court.
## Stuff i've worked on
## Recent Talks and Presentations
AI Safety in the NHS
University of York, Centre for Assuring Autonomy Safety in the NHS Symposium • 2025
Invited speaker, discussed clinical and AI research around the Dora system that's been deployed in the NHS to support patients and clinicians. Watch talk
Safe Automation of Healthcare Consultations with Dora: Conversational AI in Practice
Medical Protection Society Foundation Safety Symposium • 2025
Invited Speaker, attended by >3000 healthcare practitioners worldwide. Link
ASTRID-An Automated and Scalable TRIaD for the Evaluation of RAG-based Clinical Question Answering Systems
Association of Computational Linguistics (ACL), Main Symposium Poster, Vienna • 2025
We developed and validated a scalable and automatable safety framework to detect failure modes in a clinical RAG pipeline. This framework now guides the development of our clinical question answering modules.
Who Communicates Better? A Study on Clinician and AI-Generated Responses to Frequently Asked Patient Questions
APTOS Conference Hong Kong, Research Poster • 2024
We compared LLM versus Clinician responses to patient FAQs in cataract surgery against a suite of quantitative and qualitative metrics.
## Talk to me
I'm always interested in connecting with fellow researchers, healthcare innovators, and anyone working at the intersection of technology and medicine.