How I approached it
I worked closely with product and engineering teams to unstand the limitation, and design a conversational AI workflow that combines human input and automated reasoning.
The left panel is a conversational interface where the AI assistant gathers context, such as environment type (Production, Pre-prod, Test) and change goals.
The right panel displays the AI’s findings in a structured CI list, showing the number of affected items and allowing users to filter, review, and confirm.
To ensure consistency across AI-driven experiences, I followed the ServiceNow Horizon Conversation Design Guidelines. These principles helped define how the AI assistant communicates, handles user intent, and maintains clarity across both scripted and generated conversations.
During the design phase, we went through several rounds of prototyping and continuously aligned with the component team to ensure that the interaction patterns and system architecture worked seamlessly together.
