A pilot study of AI agent for life science research

AI agents or agentic AI are a system that augments large language models with memory, tools, and control loops. The rise of AI agents in recent years has been reshaping the imagination of how scientific research should be done. Though it is being applied across scientific research, it remains challenging to go from chatbots to reliable AI agents that can work closely with domain-specific data and tools. This pilot study aims to build the in-house capability to design, build, and evaluate an AI agent for scientific research and deliver a reusable, evidence-based guideline of how to do it well.