What are AI Memories?
AI memories are personal, contextual information stored for individual users that help your AI system:- Remember user preferences and past interactions
- Provide personalized responses based on user history
- Enhance user experience through adaptive behavior
Key Benefits
With AI memories, every interaction feels personal. The system:- Tailors responses to individual user preferences
- Anticipates needs based on past behavior
- Adapts to changing preferences over time
- Drives higher engagement through personalized experiences
- Creates a sense of connection where the AI truly “knows” the user
How It Works
1. Memory Creation
Memories can be created in two ways:- Manual Addition: Explicitly add specific memories using the Add User Memory endpoint
- Automatic Generation: Use the Generate User Memory endpoint to automatically create multiple memories from user queries or conversations
2. Memory Storage
All memories are stored in a vector database that enables:- Semantic search to find relevant memories
- Contextual retrieval based on meaning, not just keywords
- Tenant isolation to ensure data privacy and security
3. Memory Retrieval
Use the Retrieve User Memory endpoint to:- Find relevant context before generating AI responses
- Access user preferences for personalized interactions
4. Memory Management
- List all memories using the List User Memories endpoint
- Remove outdated memories with the Delete User Memory endpoint
- Maintain data quality through regular memory audits
Tenant Provisioning
Important: If you receive an error stating “Tenant-id/sub-tenant-id combination either does not exist or is not provisioned for user memory”, this means the tenant/sub-tenant combination hasn’t been set up for user memory functionality yet.
Use Cases & Scenarios
Sales Assistant
Imagine a sales assistant that remembers each rep’s preferred answer format and the types of files they use most. Over time, the AI starts responding in the rep’s favorite style and prioritizes relevant content, making every interaction faster and more effective.Customer Support
A support system that remembers each customer’s technical expertise level, preferred communication style, and past issues. This enables the AI to provide appropriately detailed responses and avoid repeating previously solved problems.Learning Platform
An educational platform that tracks each student’s learning progress, preferred learning styles, and areas of difficulty. The AI can then adapt content delivery and provide personalized study recommendations.Project Management
A project management tool that remembers each team member’s role, expertise areas, and communication preferences. This helps the AI provide relevant updates and suggestions tailored to each person’s needs.Getting Started
- Add your first memory using the Add User Memory endpoint
- Generate memories from conversations using the Generate User Memory endpoint
- Retrieve relevant context using the Retrieve User Memory endpoint
- Manage your memory profile using the List User Memories and Delete User Memory endpoints
Best Practices
- Use descriptive memory content that will be useful for future AI interactions
- Consider the context in which memories will be retrieved
- Avoid storing sensitive information unless necessary
- Use consistent formatting for similar types of memories
- Regularly review and clean up outdated memories
- Provide users with control over their memory data