Patrick Koch

In this talk, we explore how Agentic AI is reshaping Terraform workflows, moving beyond simple code generation toward intelligent, context-aware systems that can actively assist, adapt, and operate within your infrastructure lifecycle.

We will introduce the concept of personalized Copilot agents for Terraform, showing how developers can create domain-specific AI assistants that understand their infrastructure patterns, organizational standards, and preferred architectures. These agents go far beyond autocomplete: they act as collaborators in designing, validating, and evolving Terraform configurations.

Next, we will dive into Agent Skills, demonstrating how targeted capabilities can be added to these agents - enabling them to perform specialized tasks, such as:

validating Terraform plans against best practices

enforcing governance rules

assisting in refactoring and drift detection

Finally, we will explore the emerging role of MCP (Model Context Protocol) servers in Terraform ecosystems. MCP opens the door for agents to securely access external context and tools.

This enables a shift from static Infrastructure as Code to interactive, context-aware Infrastructure as Systems.

By the end of this session, you will understand:

How Agentic AI changes the way Terraform code is written and maintained

How to build and customize Copilot agents for your infrastructure workflows

How Agent Skills and MCP servers unlock powerful new capabilities

What the future of AI-driven infrastructure engineering could look like

Whether you are a platform engineer, DevOps specialist, or cloud architect, this talk will give you a practical and forward-looking perspective on the next evolution of Terraform and IaC.