AI Preferences allows you to customize how the Hatz AI responds across all your conversations and models. You can set your personal details, define the AI's communication style and tone, add custom instructions, and control AI behavior features like personalization and memory.
These preferences shape the tone, length, and formatting of AI responses. They do not change what a model can do—only how it responds.
Who this is for
This feature is available to all Hatz users who want to personalize their AI interactions. It applies at the individual user level within a workspace.
How to access AI Preferences
Navigate to Workspace in the Hatz platform
In the left sidebar under Personal, select AI Preferences
Configuration options
AI Behavior
Personalization
When enabled: Uses your memories and preferences to tailor AI responses
Toggle on or off using the switch
Memory (New Feature Coming Soon!)
When enabled: Allows the AI to remember information across conversations
Personal Default LLM
Select your Personal Default LLM will be selected when opening chat
Your personal preference takes priority over the organization default, so you always get the experience you want — no need to manually switch models every time.
A few things to keep in mind:
Your admin's policies still apply. You can only choose from models your organization has made available. Any model your admin has disabled cannot be set as a personal default.
Rate-limited models. If the model you have selected is rate-limited, Hatz will automatically fall back to your organization's default model and switch back to your selected model once it becomes available again.
Admins can change availability at any time. If your organization disables a model you've saved, your preference will be automatically cleared and you'll fall back to the organization default. You'll see a notification in Settings if this happens.
To clear your preference at any time and return to the organization default, click Clear personal default in Settings.
What is Memory?
Memory is a dynamic, auto-growing knowledge base about you that the AI builds over time during conversations.
What Memory stores:
Individual fact and summary items (e.g., "User loves pizza", "User works on a TypeScript monorepo")
Each memory has an importance score
Memories are stored and recalled based on relevance to the current conversation
How Memory works:
During conversations, the AI can automatically save, forget, search, and list memories using built-in tools
Before each AI response, the system predicts which memories are relevant (based on prior message context), fetches them, and injects them as a memory in the system message
You can manually manage memories (add, edit, delete, merge, import) in Settings > Memory
User control:
Memory is mostly automatic—the AI decides what to save based on conversation context. However, you can review and manually manage your full memory list at any time.
About you
Provide context about yourself to help the AI personalize responses:
First name: Your first name
Last name: Your last name
Timezone: Select your timezone from the dropdown
Job description: Your role or job title (e.g., "Product designer, Software engineer")
Base style and tone
Set the default style and tone for AI responses. These settings do not impact model capabilities.
Response length: (Balanced, Concise, Detailed)
Concise answers are 1–3 sentences. Balanced answers are a mix of concise and detailed. Detailed answers include full explanations and examples.
Tone: Select the tone of AI responses (Neutral, Formal, Friendly)
Emoji: Toggle emoji usage in responses (On/Off/Mirror User)
Custom instructions
Add specific behavioral, style, and tone preferences for the AI to follow across all conversations and models.
Enter your instructions in the text field (up to 2,000 characters)
Click See examples for guidance on what to include
Examples:
"Ask clarifying questions before giving detailed answers."
"Start with a short summary, then provide details."
"Use bullet points for steps and action items."
"Prioritize practical recommendations over theory."
Saving your preferences
Configure your desired settings in each section
Click Save Settings at the bottom right to apply changes
To revert all settings to default, click Reset to default
Expected results
AI responses will reflect your configured style, tone, and length preferences
The AI will use your personal details (name, job description, timezone) to provide more relevant responses
Custom instructions will be applied across all conversations and models
If Personalization and Memory are enabled, the AI will adapt to your preferences and retain context across conversations
Understanding timezone settings
The timezone you select in AI Preferences becomes your permanent setting in Hatz and is used for all time-related references in AI responses. Once saved, the AI will always reference this timezone—even if you physically move to a different location or access Hatz from a device in a different timezone. This ensures consistency across your conversations regardless of which machine or location you're using. If you relocate or want the AI to reference a different timezone, you'll need to manually update the timezone setting in AI Preferences and save your changes.
Troubleshooting
Changes aren't taking effect
Ensure you clicked Save Settings after making changes
Refresh the page and start a new conversation to see updated preferences
Not sure what to include in Custom instructions
Click See examples for guidance on effective instructions
Limitations
AI Preferences shape how the AI responds but do not change the underlying capabilities of the models
Custom instructions are limited to 2,000 characters
Changes only affect new AI interactions; existing conversation history is not retroactively modified
The specific options available for Response length, Tone, and other drop-down may vary and are limited to the preset options provided in the interface
Personalization and Memory features require those toggles to be enabled to function