Agentic AI for Autonomous Scientific Discovery

The aim of this workshop is to help participants understand how autonomous AI systems can support and accelerate scientific research by combining multi-agent intelligence, foundation models, automated experimentation, and self-driving laboratory workflows.

• Understand the concept of autonomous scientific discovery.
• Explore the role of multi-agent AI scientists in research workflows.
• Learn how foundation models support scientific reasoning and experimentation.
• Understand the basics of self-driving research labs.
• Identify applications in biotechnology, drug discovery, nanotechnology, materials science, and sustainability.
• Recognize key challenges such as reliability, ethics, and reproducibility.

📅 Day 1: Foundations of Autonomous Scientific Discovery

Topics Covered

• Introduction to autonomous scientific discovery
• From traditional research workflows to AI-powered research workflows
• Role of foundation models in scientific reasoning
• AI co-scientists and research assistants
• Literature mining, research gap identification, and hypothesis generation
• Applications in biotechnology, drug discovery, nanotechnology, materials science, and sustainability

Hands-on Activity

Google Colab Activity: Build a simple AI-powered literature insight workflow using Python to analyze research abstracts and extract keywords, research themes, and possible hypothesis ideas.

📅 Day 2: Multi-Agent AI Scientists for Research Automation

Topics Covered

• What are multi-agent AI systems?
• Role-based AI agents: literature agent, hypothesis agent, experiment planner, data analyst, and reviewer agent
• Agentic workflows for scientific problem-solving
• Prompt engineering for research agents
• Tool-using AI agents for databases, papers, code, and scientific analysis
• Reliability, validation, and human-in-the-loop research supervision

Hands-on Activity

Google Colab Activity: Create a basic multi-agent research workflow where different AI agents perform literature summarization, hypothesis generation, and experiment planning for a selected research topic.

📅 Day 3: Self-Driving Research Labs and Future of AI-Driven Science

Topics Covered

• Introduction to self-driving laboratories
• AI-driven experimental design and optimization
• Closed-loop research: plan, test, analyze, improve
• Use of machine learning for experiment recommendation
• Applications in chemistry, materials discovery, pharma, biotech, and energy research
• Ethical, reproducibility, safety, and governance challenges
• Future career and research opportunities in autonomous science

Hands-on Activity

Google Colab Activity: Simulate a self-driving lab workflow using a sample dataset where an AI model recommends the next best experiment based on previous experimental results.

• Explain how AI is transforming scientific discovery.
• Describe the use of AI agents and foundation models in research.
• Understand how self-driving labs automate experimentation.
• Identify real-world applications of autonomous research systems.
• Evaluate the benefits and limitations of AI-powered scientific workflows.