Rethinking enterprise event creation with AI
Building a whole multi-session event from a single prompt.
Problem
Learning admins at Highspot build events one piece at a time. They create an empty event, land in an overview, then set up each session separately. Same fields, again and again.
There's no view of the whole program until the end. Fine for a one-session webinar. Painful for a multi-region certification, where the same setup repeats for every session.
Approach
Our design team spent a month building on Arctic AI, Highspot's new agentic framework, each taking the area where an agent could remove the most work. I took event creation. Most of its friction is repeated setup, and that's work an agent should take over.
I didn't want the agent filling in the same forms faster. Learning admins describe the event once in chat; the agent builds the event, every session, and how they connect. Admins then finish on a canvas, dragging and opening nodes by hand, with chat open throughout. I owned scope as well as design and held one path end to end: prompt, draft, fix, publish. I prototyped it in code with Cursor instead of Figma, since an agent building an event only makes sense in motion. A simple parser stands in for the LLM.
Key decision
Once the agent drafted the event, admins needed a place to edit it. I prototyped three options. Wizard was familiar but click-heavy, and never showed the whole event. Free-form Canvas showed everything, but forced zooming, panning, and connections events don't need. I chose a simple Node Map: each part of the event is a node admins drag or open to edit. An Outline shows the same structure as a list and takes over on small screens; both open the same detail panel.
Agent drafts will sometimes be wrong, so the agent checks its own work before publish. When it finds an instructor double-booked across overlapping sessions, it proposes a new time and waits. The admin approves it or sets their own, then publishes.
Outcome
This didn't ship, and nothing was measured. The bet: setup goes from one form per session to one prompt and a review, so multi-region programs stop being the expensive ones to build. I'd test it with time to first publish and how many multi-session events get finished. Running scope myself changed how I work: I now start projects by naming the one path that has to hold up.