---
title: "Personalize Assistants"
description: "Mem0 enhances AI assistants by providing an intelligent memory layer. This allows your AI assistants to remember past interactions, user preferences, and learned information, leading to more person..."
last_updated: "2026-05-07T04:49:19.121014+00:00"
canonical_url: "https://www.doc0.dev/docs/faa36707-7c28-4f69-a18f-700ff61c704e/guide/section-3/personalize-assistants"
---

Mem0 enhances AI assistants by providing an intelligent memory layer. This allows your AI assistants to remember past interactions, user preferences, and learned information, leading to more personalized and consistent conversations. Instead of starting fresh with every interaction, your assistant can adapt to individual needs and continuously learn over time, making it ideal for applications like customer support, personalized AI assistants, and adaptive systems.

### Key Concepts

*   **Memory Layer**: This is the core component that stores and manages information for your AI assistant. It allows the assistant to recall relevant details from previous conversations or teachings.
*   **Teachability**: This feature enables your AI assistant to learn from user input and store that knowledge for future use. It can identify advice, tasks, and factual information within conversations and commit them to its memory.
*   **Memory Storage**: When an assistant is "teachable," it can automatically identify and store two types of information:
    *   **Task-Advice Pairs**: If a user asks the assistant to perform a task and provides advice on how to do it, the assistant can remember the general task and the associated advice.
    *   **Question-Answer Pairs**: If a user provides factual information, the assistant can store it as a question-answer pair, allowing it to recall that information when asked a similar question later.
*   **Memory Retrieval**: Before responding to a user, the assistant can search its stored memories for relevant information. This ensures that its responses are informed by past interactions and learned knowledge.

### Enabling Teachability for your AI Assistant

To make your AI assistant remember and learn from interactions, you can integrate the `Mem0Teachability` capability. This capability allows your assistant to automatically store and retrieve relevant information.

<Steps>
<Step>
### Set up your AI Assistant
First, ensure you have an existing AI assistant agent, typically a `ConversableAgent` from the AutoGen library, that you want to make teachable.
</Step>
<Step>
### Initialize Mem0Teachability
Create an instance of `Mem0Teachability`. You can configure its behavior using various parameters.

```python
from mem0 import Memory
from cookbooks.helper.mem0_teachability import Mem0Teachability
# Assuming 'my_llm_config' is your LLM configuration dictionary
# and 'my_agent_id' is a unique identifier for your agent.

# Optional: Initialize a Memory client if you want to reuse an existing one
# memory_client = Memory(api_key="YOUR_MEM0_API_KEY") # Or other configuration

teachability_capability = Mem0Teachability(
    verbosity=1, # Set to 1 to see when memories are stored/retrieved
    reset_db=False, # Set to True to clear all memories for this agent on startup
    recall_threshold=1.5, # Adjust how strictly memories are recalled (lower means more strict)
    max_num_retrievals=10, # Maximum number of memories to retrieve
    llm_config=my_llm_config, # Required: Your LL
</Step>
</Steps>
```


## Sitemap

See the full [sitemap](https://www.doc0.dev/docs/faa36707-7c28-4f69-a18f-700ff61c704e/llms.txt) for all pages in this wiki.