| Note: From October 1st, 2025, we introduced new plans that better reflect how our customers use Deputy: Lite, Core, and Pro. If you're an existing customer, look out for an email with all the details you’ll need. Learn more about the new plans and what this change means for you. |
| Note: Some features or functions mentioned in this article are currently in beta trials and may not be visible in your account. |
You can use Deputy AI to build schedules, edit shifts, fill open shifts, and publish schedules using natural language.
Before you read
- Target audience: This article is for users with System Administrator, Location Manager, and Supervisor access.
-
Deputy AI:
- Deputy AI is currently in beta.
- Deputy AI assists users by generating responses and proposing changes using large language models and based on information available in Deputy.
- Users are responsible for reviewing all responses and proposed changes before confirming them.
- Deputy AI does not replace human judgment and may generate inaccurate, incomplete, or inconsistent responses.
- Click to learn about how to access and navigate Deputy AI.
This article covers
- How does Deputy AI work with scheduling?
- How can Deputy AI help me schedule?
- What requests may not work well with Deputy AI?
- Troubleshooting Deputy AI
- How does Deputy AI handle my data?
- FAQs
How does Deputy AI work with scheduling?
Deputy AI helps you manage schedules using natural language while following the rules and information already configured in your Deputy account.
When you ask Deputy AI to help you complete a task, it reviews information such as team member availability, approved leave, training records, scheduling rules, and location access before generating a proposed plan for you to review.
Nothing is updated automatically. Deputy AI will only take action after you confirm the plan.
Sometimes Deputy AI may not be able to complete a request or return a result. This can happen when there are no suitable team members available, a request conflicts with scheduling rules, or more information is needed to complete the action.
How can Deputy AI help me schedule?
Deputy AI can help you manage schedules more quickly using natural language.
You can use Deputy AI to:
- create and copy schedules
- update and delete shifts
- fill and replace shifts
- publish schedules
- answer questions about staffing and availability
Note: For scheduling queries and requests, Deputy AI currently works best with up to approximately 150 shifts at a time.
| Supported request type | Description | Example requests |
|---|---|---|
| Create schedules and shifts | Create one-off, recurring, open, or rotating shifts for team members, locations, or areas. |
“Create recurring open shifts every Tuesday and Thursday from 7am–3pm for the next 6 weeks in @area.” “Create a schedule in @area for Monday to Thursday next week with three shifts each day.” “Create a 4 on 2 off 9am–5pm shift pattern for @employee in @area for the next 8 weeks.” |
| Copy schedules and shifts | Copy schedules, duplicate shifts, or repeat existing schedule patterns across future dates. |
“Copy this week's schedule for @location to next week.” “Copy @employee's shifts from 30 March to 5 April to every week for the next 4 weeks.” |
| Bulk Update schedules | Make bulk schedule updates without editing shifts individually. |
“Move all shifts scheduled in @area1 next week to @area2.” “Add a 30 min meal break to every @location shift this week.” “Change all Monday 8am shifts next week at @location to a 9am start time.” |
| Delete shifts | Remove shifts in bulk for locations, areas, date ranges, or unpublished schedules. |
“Delete all unpublished shifts at @location for this week.” “Delete all of @employee’s shifts from 1–14 September as they’re going on holiday.” |
| Fill and replace shifts | Fill empty shifts, find replacements, and reassign shifts between team members. |
“Fill all empty shifts @location for next week.” “@employee called in sick for their shift tomorrow — find a replacement.” “Reassign all of @employee1’s shifts from 1–30 June to @employee2.” |
| Swap and offer shifts | Swap shifts between employees or offer shifts to specific team members. |
“Swap @employee's Thursday shift with @employee2.” “Offer all empty shifts for next week in @area to @employee and @employee2.” |
| Publish schedules | Publish or unpublish schedules. |
“Publish next week's schedule at @location.” “Find all unpublished shifts at @location for next week and publish them.” |
| Shift notes | Add or remove shift notes. | “Add a shift note to all shifts @area for the next two weeks: ‘petrol voucher available’.” |
| Shift templates | Save or load shift templates. |
“Save this week's @location schedule as a template and call it ‘Standard Week’.” " Load the "Winter Template" for next week at @location." |
| Ask scheduling questions | Understand staffing coverage, availability, unpublished shifts, and shift allocation. |
“Who is scheduled to come in today across all my locations?” “Are there any unfilled shifts in the next 7 days across any of my locations?” “List any shifts at @location that don't have any breaks.” |
What requests may not work well with Deputy AI?
Deputy AI works best with clear scheduling requests focused on one task at a time.
-
Requests with too many actions: Deputy AI may struggle with prompts that combine several scheduling actions into a single request.
- For example: “Add shifts for next week, fill all open shifts, swap Friday shifts between John and Sarah, and then publish the schedule.”
- For example: “Add shifts for next week, fill all open shifts, swap Friday shifts between John and Sarah, and then publish the schedule.”
-
Very large scheduling actions: Deputy AI may not complete requests that span long date ranges or involve very large numbers of shifts.
- For example: “Schedule Maria every Monday and Tuesday for the next 12 months.”
- For example: “Schedule Maria every Monday and Tuesday for the next 12 months.”
-
Requests without enough detail: Deputy AI may not understand requests that are too broad or missing scheduling details such as dates, times, locations, or employees.
- For example: “Fix next week’s schedule.”
- For example: “Fix next week’s schedule.”
-
Complex scheduling structures: Deputy AI may not work well for highly detailed scheduling logic within a single shift or schedule request.
- For example: “I have four staff and need 24-hour coverage Monday to Sunday. Can you create a schedule for me?”
- For example: “I have four staff and need 24-hour coverage Monday to Sunday. Can you create a schedule for me?”
-
Staffing optimisation and forecasting: Deputy AI cannot automatically optimise schedules based on labour costs, sales forecasts, or historical staffing trends.
- For example: “Build a schedule that optimises my labour costs.”
Troubleshooting Deputy AI
If Deputy AI is not responding as expected, try these steps:
-
Click New Topic: If Deputy AI seems confused or is referencing old data, click New Topic to clear the conversation context.
-
Be more specific: To ensure an accurate request, always include relevant information like dates and times, and use the @ symbol to identify and match names, locations, and areas in your request.
- Refresh the page: If the action returns as complete but the schedule or timesheets do not update, refresh the page and check again.
- Shorten your prompt: If an action fails or produces inconsistent results, try a shorter prompt, remove unnecessary detail, or break the request into smaller actions.
- Review the relevant employee profile: If issues occur with a specific team member, review their profile for potential scheduling, availability, employment, or configuration issues.
- Check permissions: If you receive an error when clicking Confirm, make sure you have permission to perform that task in Deputy.
How does Deputy AI handle my data?
Deputy AI is built with security and privacy as key priorities.
- No model training: Your customer data is not used to train the large language models that power Deputy AI.
- Permission aware: Deputy AI respects existing role permissions and cannot perform actions that a user is not permitted to do manually.
- Bias protection: Guardrails are designed to reduce the risk of inappropriate or unfair recommendations. However, all recommendations should be reviewed by users before confirmation.
To turn off Deputy AI for your business, follow these instructions.
FAQs
To learn more about Deputy AI's privacy FAQs visit the Deputy Privacy Center, and to learn more about Deputy's security program visit the Deputy Trust Center.