Concierge isn’t just smart—it’s context-aware, multi-skilled, and built to handle whatever you throw at it. Here’s how it understands your questions and gets things done faster than ever.

How does Concierge work?

Concierge is a hybrid AI agent built on top of GenAI, automation tools, and a powerful rules engine. It figures out what you’re trying to do, pulls in the right tools, and responds with answers—or actions.

If it doesn’t know something? It escalates to a real person.

What makes Concierge intelligent?

Concierge uses a few smart tricks to figure out how to help:

• Context-aware – It knows where you’re asking, so it can tailor the response (more below).
• Multi-modal – It acts like different assistants (support agent, expense assistant, etc.) all in one conversation.
• Hybrid support – It combines AI + humans, switching seamlessly when needed.
• Natural language understanding – Just talk to it like you would to a coworker.

What is contextual behavior?

Concierge adapts based on where and how you ask questions. That means:

• If you chat in a report, it answers about that report.
• If you’re in a DM, it considers all your expenses.
• If you ask from a Workspace chat, it focuses on that Workspace.
• If you’re on a specific expense, it assumes you’re talking about that expense.

You don’t have to explain every detail—just speak naturally and Concierge will fill in the blanks.

What is a multi-modal agent?

Instead of having multiple bots for different things (like one for expenses and another for support), Concierge is all-in-one.

That means:

• You can ask anything—no need to figure out who to ask.
• You can combine requests in one message.
  • Example: “Is this reimbursable? Can you add a $12 taxi expense too?”

Can I ask multiple things at once?

Definitely. Concierge can understand and respond to complex or multi-step questions—as long as it’s clear what you want.

Here’s what works well:

• “Create a $5 lunch expense and add it to my April report.”
• “What does ‘non-reimbursable’ mean, and can you mark this expense as such?”

The key is being specific. The clearer you are, the better Concierge can help.

What happens when Concierge doesn’t know something?

If the AI hits a wall, it won’t leave you hanging. It’ll bring in a real person—often without you needing to ask.

You can also say “Talk to a human” at any time, and Concierge will escalate the conversation.

What powers Concierge behind the scenes?

Here’s what fuels its intelligence:

• A custom-trained GenAI model
• A rules engine that handles logic and automation
• Real-time access to your Expensify data (just enough to help)
• Escalation paths for live support when needed

TL;DR — What makes Concierge different?

• Knows where you’re chatting and tailors responses
• Understands natural, multi-part questions
• Does support, expenses, and reporting in one place
• Switches to a human when it gets stuck
• Just works—no setup, no training required