1 minute read

Picture: SciLifeLab’s agentic AI workshop
Last month, I attended the “Developing AI Agents in Life Sciences” workshop — a focused, hands-on event exploring how autonomous AI systems are transforming research in the life sciences.

A warm thank you to Mahbub Ul Alam, Dinh Long Huynh, and Johan Alfredéen for leading the workshop and sharing their insights. Their expertise made the sessions both technically rich and highly engaging, and the enthusiasm from participants demonstrated the growing interest in agentic AI across the life science community.

Key Topics from the Workshop

AI Agents in Life Science Research

The workshop began with an overview of how AI agents can support and automate scientific workflows. These systems can integrate diverse data sources, run predictive or analytical models, and coordinate multi-step experimental procedures. This emerging approach has the potential to accelerate discovery and reduce manual bottlenecks in research.

Building AI Agents in Practice

Participants developed their own AI agents using LangChain and the ReAct reasoning pattern. A drug discovery use case demonstrated how agents can assist with tasks such as molecular property prediction, hypothesis generation, and literature analysis.

Agent Collaboration and Interoperability

A central theme was the Model Context Protocol (MCP), an emerging standard designed to connect AI agents with external tools, data sources, and research platforms. The protocol enables interoperability and extensibility, and it can be used to integrate with resources available through SciLifeLab Serve.

Deploying AI Agents in Research Environments

The final discussions covered key considerations for real-world deployment, including:

  • Infrastructure and technical requirements
  • Data governance and legal frameworks
  • Security and ethical considerations
  • Practical strategies for integrating agentic workflows into existing research environments

Stay Connected and Explore More

  • Workshop slides and tutorials and different links: github

  • Join the SciLifeLab AI Network (Slack): https://lnkd.in/d_JwyShu
  • Explore tools for AI/ML in life sciences: https://lnkd.in/dVUYDcKU
  • Follow the SciLifeLab AI Seminar Series: https://lnkd.in/d9fcBny8
  • Browse research tools available through SciLifeLab Serve: https://lnkd.in/dwJEg82m

Post credit: scilifelab data centre

Leave a comment

You may also enjoy