Resources
Event
AI
On Demand

Putting Agentic AI to Work

October 23, 2026
Speakers
Justin Garrity
Vice President, Product
Josie Sutcliffe
VP of Marketing
ChartHop
Summary

See AI agents do real HR work.

Agentic AI was everywhere at HR Tech this year. In our live poll, more than half of attendees said they're thinking about AI agents but aren't using them yet. In this third session of our AI in HR and People Ops series, we show what putting one to work looks like.

Working in a demo company, Justin Garrity, ChartHop's VP of Product and Design, walks through the building blocks behind agentic actions: people and jobs, fields and forms, and actions that run on an event or a schedule. Then he builds two agentic actions live in Ask ChartHop just by describing them, and gives a first look at the prompt library, a set of ready-made prompts for HR's most useful cases.

Part of a four-session series

This is Session 3 of 4. Session 4, Beyond HR: agents across your business, is on Friday, November 6 with Ian White, ChartHop's CEO. Register once to join the final session live, or catch each recording here afterwards.

Key Takeaways
  • Describe the outcome, skip the logic.
    Traditional automation builders make you think through every if-then rule and condition. With agentic actions, you write what you want in plain language and the AI agent works out how to do it.
  • Agentic actions build on blocks you already have.
    People and jobs, fields and forms, and actions that run on an event, like a new hire, or a schedule, like every Monday. An agentic action puts an AI prompt inside that action instead of a fixed message.
  • Let the agent ask the follow-up questions.
    When Justin asked for a weekly briefing, Ask ChartHop asked one question at a time: which day, which channel, which ATS, how much cost detail. That's easier than facing a long form with every option at once.
  • Test it before it runs.
    Justin ran a test of the briefing and checked the activity log to see exactly what it would send by email and Slack: goal progress, risks, shout-outs for strong performers, and attrition risk with suggested next steps.
  • Start from the prompt library.
    A blank chat box is the hardest place to start. The prompt library groups ready-made prompts by tag, including agentic actions that combine people and business data, like pushing approved roles to the ATS and notifying finance of the cost.
Agenda
  1. Where HR teams are with AI agents today (live poll)
  2. The building blocks: people and jobs, fields and forms, and actions
  3. From actions to agentic actions: putting an AI prompt inside an automation
  4. Live build: a weekly team briefing on goals and attrition risk, created in Ask ChartHop
  5. A first look at the prompt library
  6. Live build: push approved roles to the ATS and notify finance of the cost

What’s new at ChartHop