
Clay Townsend

Elijah Davis
Clear skill boundaries and separation of responsibilities
Validation loops and drift detection
Approval gates and human-in-the-loop controls
Policy enforcement and guardrails
You're all set, [Name].
Registration closed.














Natural language is becoming a control plane for Infrastructure-as-Code.
Join Fivetran and customer Inova, Northern Virginia's leading nonprofit healthcare provider, live on March 27 for a technical deep dive into IaC at scale using MCP and agent skills. Inova's Principal AI and Data Solutions Architect, Clay Townsend, will share how their team approaches automated workloads and infrastructure at scale, and how they adopted Fivetran to support those initiatives.
You’ll see how an agent orchestrator turns intent into an executable workflow by delegating to specialized MCP servers and skills across metadata, secrets, configuration, and Terraform—while preserving control, safety, and repeatability. We’ll also break down the core design patterns behind agent-driven infrastructure and map them to a practical reference workflow: request → plan → deploy → verify.
You’ll leave with a clear mental model for building repeatable automation where agents coordinate workloads reliably, reduce manual effort, and accelerate time to value — while maintaining governance.
Clear skill boundaries and separation of responsibilities
Validation loops and drift detection
Approval gates and human-in-the-loop controls
Policy enforcement and guardrails
End-to-end auditability

Clay Townsend

Elijah Davis

Clay Townsend

Elijah Davis
Natural language is becoming a control plane for Infrastructure-as-Code.
Join Fivetran and customer Inova, Northern Virginia's leading nonprofit healthcare provider, live on March 27 for a technical deep dive into IaC at scale using MCP and agent skills. Inova's Principal AI and Data Solutions Architect, Clay Townsend, will share how their team approaches automated workloads and infrastructure at scale, and how they adopted Fivetran to support those initiatives.
You’ll see how an agent orchestrator turns intent into an executable workflow by delegating to specialized MCP servers and skills across metadata, secrets, configuration, and Terraform—while preserving control, safety, and repeatability. We’ll also break down the core design patterns behind agent-driven infrastructure and map them to a practical reference workflow: request → plan → deploy → verify.
You’ll leave with a clear mental model for building repeatable automation where agents coordinate workloads reliably, reduce manual effort, and accelerate time to value — while maintaining governance.

