Notes on RPA, agentic AI, and ServiceNow — mostly the operational lessons that don't show up in the docs.
Traditional ITSM triage relies on rigid categorization rules. An agentic approach can read the actual request and route it correctly — with the right guardrails.
Most Virtual Agent deployments fail at the knowledge layer, not the conversation layer. Pairing it properly with AI Search fixes the actual bottleneck.
Giving an AI agent the ability to take real actions in enterprise systems means governance can't be an afterthought. A practical framework for guardrails that don't kill velocity.
The mechanical steps of building an RPA bot are easy to find. The judgment calls that determine whether it survives contact with production aren't.
IntegrationHub makes it easy to wrap any REST API as reusable flow actions. That ease is also how teams end up with a dozen redundant, half-maintained spokes.
Why I built this site the way I did, and what's coming next.