# Shape Their Persona

Each employee has an identity you set on their Profile tab, and a brain the whole workspace shares from one place. This article covers both: who the employee is, and how the AI behind every employee is configured.

## TL;DR

Go to **Workspace**, select an employee, open the **Profile** tab to set their name, role, background, language, and tone of voice. The AI model, thinking depth, and actions-per-turn limit are workspace-wide settings, not per-employee: change them at **Settings > AI Models**. Profile fields save automatically when you click away.

## What Is on the Profile Tab

The Profile tab is divided into four sections:

### Identity

Controls who the employee is.

| Field      | What It Does                                                                                                |
| ---------- | ----------------------------------------------------------------------------------------------------------- |
| **Avatar** | The employee's profile picture. Click it to upload a photo (JPEG, PNG, GIF, or WebP, up to 5 MB)            |
| **Name**   | Display name used in the sidebar, chat, and across the platform                                             |
| **Role**   | Job title shown on the employee's profile (e.g., "Content Writer," "Marketing Strategist")                  |
| **Team**   | Which team the employee belongs to. Changing it moves them the same way dragging them on the org chart does |

Reports To and Mentor are shown as read-only info: Reports To is derived from the employee's team leader (or you, if there is no leader), and Mentor is the same guide for every employee.

### Persona

Controls who the employee is, not how they talk.

| Field                        | What It Does                                                                                                         |
| ---------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| **Background & Personality** | A text field where you describe the employee's background, what they care about, and how they think about their work |

This is the who, not the how, tone and language live in the Voice section below and should not repeat here. Each employee starts with a template-seeded default, and the employee can refine its own persona over time through the same field.

### Voice

Controls how the employee talks.

| Field             | What It Does                                                                                             |
| ----------------- | -------------------------------------------------------------------------------------------------------- |
| **Language**      | The employee's default language. Leave it empty and the employee mirrors whichever language you write in |
| **Tone of Voice** | Free text describing tone, e.g. "Warm and direct. Plain English, no jargon."                             |

### Contact

Where people can reach this employee directly.

| Field     | What It Does                                                                                                                                               |
| --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Email** | A live mailbox address (`<employee>@<yourworkspace>.mail.sista.ai`). A public employee replies from the same address in an isolated external conversation. |
| **Slack** | Shows the connected Slack handle, or a button to connect Slack for this employee                                                                           |
| **Phone** | Not available yet                                                                                                                                          |

## AI Models Settings

These are not per-employee settings today. They are set once for the whole workspace at **Settings > AI Models** (`/technical/ai-models`), and every employee uses the same brain.

| Field                | What It Does                                                                                                              |
| -------------------- | ------------------------------------------------------------------------------------------------------------------------- |
| **Model**            | Which AI model every employee uses by default. Grouped by tier: Fast, Standard, Advanced, and Frontier                    |
| **Credits Saver**    | A toggle that automatically routes simple, conversational turns to a faster model and keeps the model above for real work |
| **Thinking Depth**   | How hard the model reasons before responding. Only active for models that support extended reasoning                      |
| **Creativity**       | The model's temperature (0 to 2, default 0.7). Lower is steadier and more predictable, higher is more varied              |
| **Actions per Turn** | How many actions (tool calls, reads, searches) an employee can take in a single response                                  |

The model dropdown shows models from OpenAI, Anthropic, and other providers, grouped by tier. Bigger models are smarter but slower and cost more credits per turn.

### AI Model Tiers

| Tier         | Best For                                  |
| ------------ | ----------------------------------------- |
| **Fast**     | Quick responses, lowest credit cost       |
| **Standard** | Good balance of speed and capability      |
| **Advanced** | Best quality, highest credit cost         |
| **Frontier** | Top-of-line brains, the very best quality |

Exact credit cost per model changes as providers change their pricing, so the number shown next to each model in the picker is always the current one, not a fixed table.

### Thinking Depth

For models that support extended thinking:

| Level                               | What It Does                                                            |
| ----------------------------------- | ----------------------------------------------------------------------- |
| **Auto (Recommended)**              | Adjusts dynamically: quick for simple messages, deeper for complex ones |
| **Off**                             | No reasoning. Fastest responses, lowest cost                            |
| **Minimal / Light / Medium / Deep** | Increasing levels of reasoning depth and cost                           |

The dropdown is disabled when the selected model does not support reasoning.

## What You Can Do

| Action                                | How                                                                    |
| ------------------------------------- | ---------------------------------------------------------------------- |
| **Change the name**                   | Profile tab, Identity section. Edit the Name field, click away to save |
| **Change the role**                   | Profile tab, Identity section. Edit the Role field, click away to save |
| **Upload a photo**                    | Profile tab, Identity section. Click the avatar                        |
| **Move the employee to another team** | Profile tab, Identity section. Change the Team field                   |
| **Set background and personality**    | Profile tab, Persona section. Write what shapes who they are           |
| **Set language and tone**             | Profile tab, Voice section                                             |
| **Switch the workspace AI model**     | Settings > AI Models. Select from the Model dropdown                   |
| **Set thinking depth**                | Settings > AI Models. Select from the Thinking Depth dropdown          |
| **Set actions per turn**              | Settings > AI Models. Enter a number for the workspace                 |

## How to Set It Up

1. Go to **Workspace**, click on an employee in the sidebar
2. Open the **Profile** tab
3. Edit Identity, Persona, or Voice fields. Changes save automatically when you click away
4. To change the AI model or reasoning depth for every employee, go to **Settings > AI Models**

## Tips and Tricks

- **Match the persona to the role.** A customer-facing writer might be friendly and conversational. A legal analyst might be formal and precise.
- **Keep tone and language in Voice, not Persona.** Persona is background and personality; repeating tone instructions there just duplicates what Voice already handles.
- **Use Fast for simple tasks.** Quick lookups, formatting, and drafts cost less than Standard, Advanced, or Frontier models.
- **Try Credits Saver.** It automatically sends simple, conversational turns to a cheaper model and keeps your chosen model for real work, at no setup cost.
- **Persona is not a skill.** Do not put process instructions in the persona field. Use the Skills tab for workflows and the Duties tab for standing rules.

## Good to Know

- **Model, thinking depth, and actions per turn are workspace-wide.** Every employee currently shares the same brain settings. There is no per-employee override in the UI today.
- **Changes apply immediately.** Updated persona, voice, and AI settings take effect on the next message. No restart needed.
- **No quality impact from persona.** Persona and Voice change style, not the capability or accuracy of the work.
- **Provider fallback.** If a model provider goes down, the gateway automatically retries with a model from another provider, and the request is billed at the fallback model's rate.
- **Auto-save on blur.** Profile fields save automatically when you click outside (blur). There is no save button.
- **Change history.** The bottom of the Profile tab shows who changed the employee's details and when, since the employee itself can also rewrite its own persona.

## Frequently Asked Questions

**Q: Does changing the personality affect the quality of work?**
A: No. Persona changes the style of communication (background, personality) but not the quality or accuracy of the output.

**Q: Can I set a different AI model for each employee?**
A: Not today. The AI model, thinking depth, and actions per turn are set once at Settings > AI Models and apply to every employee in the workspace.

**Q: What happens if the AI provider goes down?**
A: The gateway automatically retries with a fallback model from a different provider. You are billed at the fallback model's rate, not the original.

**Q: Can I change the employee's name?**
A: Yes. Edit the Name field in the Identity section of the Profile tab. The change takes effect immediately.

**Q: What does "Actions per Turn" mean?**
A: It limits how many actions (tool calls, reads, searches) an employee can perform in a single response. Higher values allow longer, multi-step work.

**Q: Can I upload a custom photo for my employee?**
A: Yes. Click the avatar on the Profile tab and upload a JPEG, PNG, GIF, or WebP file up to 5 MB.