A timely introduction to AI-assisted coding in healthcare
GCLS AI Lab partnered with the Data Science and Health Research Excellence Cluster at the University of British Columbia (UBC DASH) for Vibe Coding for Healthcare 101, a webinar series. It ran from 15 April to 13 May 2026.
Across three interactive sessions, participants learned what vibe coding is, why it matters in healthcare, and how natural language can be used to build functional health software without traditional programming experience.
From a natural-language idea to a working health tool
The series focused on practical work rather than abstract demonstrations. Participants used AI tools including Replit and Claude Code/Cowork to move from a description to a working prototype, then considered how research automation and clinical workflows could benefit from the same approach.
- Explain what vibe coding is and why it matters for health research and clinical practice
- Use Replit and Claude Code/Cowork to build a functional health tool from a natural-language description
- Build a basic research automation workflow for literature screening, data extraction, or synthesis
- Identify appropriate use cases and understand where compliance and ethics considerations apply
- Identify next steps through structured learning pathways
A bridge from curiosity to responsible practice
The partnership showed how accessible AI-assisted coding can help researchers and clinicians test ideas faster while keeping use cases, compliance, and ethics in view. It is an invitation to begin with a real problem, build something functional, and deepen the work through structured learning.
All videos and resources are available on the specially developed microsite, including recordings from each session.
Common questions
What is vibe coding?
Vibe coding is an AI-assisted way to build software by describing what you want in natural language and collaborating with coding agents to produce a functional tool.
Who was the webinar series for?
It was designed for people working in health research and clinical practice who wanted to explore practical AI-assisted software building without traditional programming experience.
What tools were covered?
The sessions included Replit and Claude Code/Cowork, alongside discussion of research automation, clinical use cases, compliance, and ethics.
Sources and further reading
Company pages describe the published product model. Research and vendor references provide context, not independent validation of Perceptors.
