VTK Agents: Making Scientific Software More Accessible with AI

Complex scientific software can be difficult to learn, especially for new developers. The Visualization Toolkit (VTK) is a widely used open source framework for scientific visualization, but its steep learning curve can make getting started challenging.

Large language models (LLMs) and AI agents offer new ways to address these challenges. By integrating AI into existing development workflows, teams can make complex software more approachable while supporting ongoing maintenance.

In this webinar, Kitware developers Vicente Bolea and Jeff Lee will share their experiences integrating LLMs and agentic AI into VTK. They will explore how these technologies can improve accessibility, simplify development workflows, and help maintain large scientific software projects.

During this webinar, we’ll explore:

• Lessons learned from integrating LLMs and agentic AI into VTK’s development workflows.
• An introduction to vtk-prompt and vtk-prompt-ui, including how these tools support accessibility.
• An overview of the vtk-intelligence ecosystem.
• Specialized AI-powered tools designed to assist with VTK maintenance.

The session will provide practical insights into applying AI agents to established open-source software projects.

Key Takeaways

• Understand how LLMs and agentic AI can support large open-source software projects.
• Learn how AI-powered applications can make complex scientific software more accessible.
• Explore tools and approaches that can simplify software development workflows.
• Discover how AI agents can assist with the ongoing maintenance of scientific software.

Who Should Attend

• Software developers working with open source projects.
• Project leaders managing large software projects.
• Researchers and engineers interested in scientific software.
• Professionals interested in software accessibility.
• Individuals working with continuous integration (CI) and software development workflows.