Imagine a chatbot makes a joke, and the user gives it a negative feedback. The same agent is far funnier on the said topic a week later. This is what we mean by agent\u2019s live adaptation. Most chat agents as of now are static programs in nature. They don\u2019t really learn from user feedback during runtime. But, we can implement the critical ability of running time learning using an advanced system. This tutorial provides a description of how to perform continuous learning using the AutoGen teachability capability. We will also give a runtime demo and good ways to achieve successful agent adaptation.\u00a0<\/p>\n
What is AutoGen?<\/h2>\n
AutoGen is a powerful framework to create complex AI workflows. It is excellent in its ability to orchestrate conversations between several autonomous agents. This multi-agent framework gives each of their agent\u2019s different roles. Agents collaborate towards complex and multi-step task solutions.\u00a0<\/p>\n
The system uses composable conversation patterns, and the user determines the agents and their communication protocol. This design does not have the same limitations associated with single prompt AI. AutoGen supports various types of agents like AssistantAgent and UserProxyAgent. In this setting, teachable AI agents<\/a> can be invented in an ideal environment. The framework is extensible, making it a good fit to provide new capabilities, including learning. <\/p>\n Read more: Build Multi-Agent Frameworks using AutoGen<\/a><\/em><\/p>\n An agent learns from interactions in terms of changing its future behavior. This change is the outcome of experiences that occurred during a live conversation. In offline fine-tuning, the model trains on a massive, fixed dataset and needs a long time to finish. Here, you focus on making local, immediate, high-impact corrections.<\/p>\n Teachable AI agents need to alter their responses right once. They can learn to fix a fact or change a persona and then quickly change their tone, preference, or strategy for a task. This depends on the ability of the agent to store context. The ability to learn from the feedback generated by the users is closely bound with the utility of the system. <\/p>\n Within this multi-agent framework, the teachability capability can be used to provide the foundation of an adaptable agent. This is a utility that can be used to store and retrieve knowledge gained in a conversation. AutoGen teachability capability does the heavy lifting to create the mechanism for persistent learning.\u00a0<\/p>\n This ability works by vectorizing conversation snippets and user corrections and storing this in a local vector database. As such, this database represents a persistent agent of memory. Then for any new question coming, the agent simply checks its memory, retrieves relevant past lessons\/corrections and injects them into its current prompt and lets the model instantly adapt its response. This is an on-demand adaptation with no API call or updating model weights. Through this process, authentically adaptive and responsive agents are produced.<\/p>\n The architecture for our learnable agent is still very simple and consists of regular components found in the AutoGen toolkit; two main agents and a single core capability.\u00a0<\/p>\n The The key components include:\u00a0\u00a0\u00a0<\/p>\n This simple yet powerful set up gives the foundation to advanced teachable AI agents.\u00a0<\/p>\n We use the code provided in the Python<\/a> notebooks to implement our learnable comedian. This hands-on process clearly demonstrates the configuration. We first install required dependencies of AutoGen and teachable components of AutoGen.\u00a0<\/p>\n Next, we define the configuration followed by the agents.\u00a0<\/p>\n Define Configuration and Agent Setup<\/strong>\u00a0<\/p>\n First, we define the large language model configuration list. It contains the model and API key of the agent. The Attach the Teachability Capability<\/strong>\u00a0<\/p>\n This step introduces the AutoGen teachability capability. We instantiate Initiate the Chat Loop<\/strong>\u00a0<\/p>\n We create the The notebook\u2019s output clearly demonstrates the learning process. The agent starts with a conventional attempt.\u00a0<\/p>\n The agent attempts to make a third joke that is more descriptive and relatable. It has captured every explicit piece of user input and stored the dislikes in its memory for future use. Based on that memory, it will try to avoid such jokes in later conversations. The system is dynamic and adapts to the humor profile of the user in real-time. The capability to learn immediately from feedback from users is what distinguishes this agent as a dynamic agent.<\/p>\n A successful teachable agent generates measurable improvements in performance. The utility and accuracy of adaptation must be the subjects of evaluation. We use short-term and long-term measures for assessment.\u00a0<\/p>\n Short-term measures provide immediate success. They are measurements of appropriateness of response after correction. They assess whether an agent did what was stipulated in the direct feedback instruction. Long term metrics measure knowledge of retention. They measure the decrease in the repetition of mistakes in many sessions. A\/B testing<\/a> might be used to compare the performance of teachable AI agents<\/a> with non-teachable baselines. Safety monitoring is of great importance at the same time. We need to make sure that we don\u2019t have any unsafe or biased outputs from the agent that is learning from user interactions.\u00a0<\/p>\n A design consideration for any learning loop must emphasize stability. PK teachable AI agents need to integrate carefully with existing systems.<\/p>\n The implementation of the AutoGen teachability capability holds a number of risks. Active mitigation by the developers is required against these common pitfalls.\u00a0<\/p>\n The basic teachable agent is a foundation for more complex systems. The multi-agent framework supports sophisticated learning architecture.\u00a0<\/p>\n One advanced concept is the hybrid approach, where fast, transient memory combines with scheduled offline model fine-tuning<\/a>. This achieves the best of both worlds: immediate reaction and deep, long-term improvement.<\/p>\n Another avenue that can be taken involves multi-agent learning. In this scenario, agents teach each other in a collaborative environment. An agent shares corrections with a whole group of workers. This idea expands the core concept beyond one single agent.<\/p>\nWhat \u201cLearn from Interactions\u201d Means?<\/h2>\n
Key Concept: Teachability Capability<\/h2>\n
Architecture and Components<\/h2>\n
UserProxyAgent<\/code> and the AssistantAgent<\/code> are the discussion participants. The AssistantAgent comes with a particular person- the comedian, for example. AutoGen teachability capability is an attachment that is directly attached to the AssistantAgent. It includes the logic to store and retrieve lessons. With this configuration, the agent has a practical persistent agent memory store.\u00a0<\/p>\n\n
Step-by-step Implementation Walkthrough<\/h2>\n
!pip install autogen ag2[teachable]\u00a0<\/code><\/pre>\nAssistantAgent<\/code> is our comedian, who has a certain persona defined.\u00a0<\/p>\n# Define the configuration list with environment variables\u00a0\n\nconfig_list = [\u00a0\n\n\u00a0\u00a0\u00a0{\u00a0\n\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\"model\": \"gpt-4o-mini\",\u00a0\n\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\"api_key\": userdata.get('OPENAI_KEY')\u00a0\n\n\u00a0\u00a0\u00a0}\u00a0\n\n]\u00a0\n\n# Create an instance of AssistantAgent for a comedian\u00a0\n\ncomedian = AssistantAgent(\u00a0\n\n\u00a0\u00a0\u00a0name=\"comedian\",\u00a0\n\n\u00a0\u00a0\u00a0system_message=\"You are a professional comedian. You can tell jokes and entertain people.\",\u00a0\n\n\u00a0\u00a0\u00a0llm_config={\"config_list\": config_list}\u00a0\n\n)<\/code><\/pre>\nTeachability<\/code> and attach it to the comedian agent. The path_to_db_dir<\/code> sets up the local storage for the persistent agent memory. We set reset_db=False<\/em> so that the agent retains knowledge across runs.\u00a0<\/p>\nfrom autogen.agentchat.contrib.capabilities.teachability import Teachability\u00a0\n\n# Create an instance of Teachability\u00a0\n\nteachability = Teachability(\u00a0\n\n\u00a0\u00a0\u00a0reset_db=False,\u00a0\n\n\u00a0\u00a0\u00a0path_to_db_dir=\".\/comedian_assistant_experience\",\u00a0\n\n\u00a0\u00a0\u00a0llm_config={\"config_list\": config_list}\u00a0\n\n)\u00a0\n\n# Add the Teachability capability to the comedian agent\u00a0\n\nteachability.add_to_agent(comedian)<\/code><\/pre>\nUserProxyAgent<\/code>, which will handle the interface. Now, we begin a chat to test the teachable AI agents, PK. The chat loop records the input, output, and all user feedback. This process allows the agent to immediately learn from user feedback\u00a0\u00a0<\/p>\n# Create an instance of UserProxyAgent\u00a0\n\nuser_proxy = UserProxyAgent(\u00a0\n\n\u00a0\u00a0\u00a0name=\"user_proxy\",\u00a0\n\n\u00a0\u00a0\u00a0code_execution_config={\"work_dir\": \"coding\", \"use_docker\": False}\u00a0\n\n)\u00a0\n\n# Initiate a chat between the user_proxy and the comedian agent\u00a0\n\nuser_proxy.initiate_chat(\u00a0\n\n\u00a0\u00a0\u00a0comedian,\u00a0\n\n\u00a0\u00a0\u00a0message=\"Tell me a joke about cats and ninjas.\"\u00a0\n\n)<\/code><\/pre>\n\n
![\"Initial](\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/11\/image4-7.webp\")
<\/figure>\n<\/div>\n\n
![\"Secondary](\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/11\/image5-7.webp\")
<\/figure>\n<\/div>\n![\"Final](\"https:\/\/cdn.analyticsvidhya.com\/wp-content\/uploads\/2025\/11\/image6-3.webp\")
<\/figure>\n<\/div>\nEvaluation: How to Measure Learning<\/h2>\n
Best Practices and Practical Tips<\/h2>\n
\n
Common Pitfalls and How to Avoid Them<\/h2>\n
\n
Extensions and Advanced Ideas<\/h2>\n
![\"Harsh](\"https:\/\/av-eks-lekhak.s3.amazonaws.com\/media\/lekhak-profile-images\/converted_image_0fBqNLi.webp\")
<\/p>\n