> Source: https://builder.ema.ai/v2/templates/it-helpdesk-ai-employee
> Title: IT Helpdesk AI Employee

# IT Helpdesk AI Employee

The IT Helpdesk AI Employee is built from the **IT Helpdesk** curated template (category: IT & Ops). It is a chat AI Employee (AIE) that triages IT support requests: it classifies the ticket by category and priority, searches your runbooks for matching troubleshooting steps, and then either provides self-service resolution steps or recommends escalation to the right team. It resolves common issues instantly and routes the rest consistently, so employees get faster resolution and IT staff focus on complex infrastructure work.

## Workflow

The template ships a three-agent workflow that runs left to right:

Node

Agent type

What it does

Classify Ticket

`intent_classification`

Sorts the request into a category and assigns a priority (P1–P4). Ships with six categories: `password_reset`, `software_install`, `hardware`, `network`, `access_permissions`, `other`.

Search Runbooks

`search_respond`

Searches your IT runbooks and documentation for step-by-step resolution procedures matching the classified issue.

Resolve or Escalate

`custom`

Decides whether to provide self-service resolution steps or recommend escalation. Gives self-service instructions for password resets and common software issues; recommends immediate escalation for P1/P2 hardware or network issues.

A `start` node feeds the IT request into the graph, and a `publish` node exposes the resolution or escalation recommendation. The edges run `start → classify → search → resolve → publish`.

The Classify Ticket node ships with its categories pre-filled in the agent's `type_config`. For example:

```json
{
  "intents": [
    { "label": "password_reset",    "description": "Password reset or account unlock requests",
      "examples": ["I forgot my password", "My account is locked"] },
    { "label": "software_install",  "description": "Software installation, updates, or license requests",
      "examples": ["I need Zoom installed", "Can I get a Figma license?"] },
    { "label": "hardware",          "description": "Hardware issues, replacements, or new equipment",
      "examples": ["My monitor is broken", "I need a second monitor"] },
    { "label": "network",           "description": "WiFi, VPN, or connectivity issues",
      "examples": ["VPN is not connecting", "WiFi is slow in the conference room"] },
    { "label": "access_permissions","description": "Access to systems, folders, or tools",
      "examples": ["I need access to the shared drive", "Can I get admin rights?"] },
    { "label": "other",             "description": "Other IT requests",
      "examples": ["How do I connect to the printer?", "My badge is not working"] }
  ]
}
```

## Starting configuration

The template clones in with this configuration, which you can change in the AI Employee builder:

-   **Model selection** — Let Fusion Pick (EmaFusion™ chooses the model per request).
-   **Optimization priority** — Fastest.
-   **Conversation context** — enabled, so the AI Employee retains context across turns.
-   **Feedback** — enabled, with thumbs-up / thumbs-down reasons and free-text comments.
-   **Welcome message** — "Hi! I'm your IT support assistant. Describe your issue and I'll help resolve it."
-   **Conversation starters** — three buttons: "Password reset", "Software install", "Network issue".

**Recommended data sources:** connect your IT knowledge base, troubleshooting runbooks, and software inventory documentation so the agent can resolve issues accurately.

## How to use

After you select **Use template** and the AI Employee opens in the builder:

1.  **Connect your runbooks.** Add your IT runbooks and documentation to the Search Runbooks node. See [Knowledge bases](/builder/v2/core-concepts/knowledge-bases).
2.  **Customize categories and priorities.** Edit the ticket categories and priority levels in the Classify Ticket node to match your taxonomy.
3.  **Set escalation rules.** Edit the Resolve or Escalate node's Instructions to define which teams handle which categories and which conditions trigger escalation.
4.  **Optionally connect Tools.** Add Tools for automated remediation — for example, password resets or software provisioning. See [Tools](/builder/v2/agent-reference).
5.  **Publish and deploy.** Publish the workflow, then connect it to your ticketing or chat platform. See [Channels and deployment](/builder/v2/integrations-data).

> [INFO]
> **Resolution and escalation are LLM-driven by default.** Out of the box, the Resolve or Escalate node recommends self-service steps or an escalation path; it does not perform remediation on its own. To let it take action — resetting a password, provisioning software — connect the relevant Tools in the builder.

## Example queries

-   "I'm locked out of my account after too many login attempts."
-   "VPN won't connect from home since this morning."
-   "My laptop won't charge and I have a customer demo in an hour."

## Related

-   [Agent reference](/builder/v2/agent-reference) — the [Intent Classifier](/builder/v2/agent-reference/intent-classification), [Search & Respond](/builder/v2/agent-reference/search-and-respond), and [Custom](/builder/v2/agent-reference/custom) agent types, and connecting Tools.
-   [Knowledge bases](/builder/v2/core-concepts/knowledge-bases) — connecting your IT runbooks.
-   [Templates overview](/builder/v2/templates) — the full curated template list and the clone flow.
