ai-agents-for-beginners

Esplorare i framework per agenti AI

(Clicca sull’immagine sopra per guardare il video di questa lezione)

Explore AI Agent Frameworks

AI agent frameworks sono piattaforme software progettate per semplificare la creazione, il deployment e la gestione di agenti AI. Questi framework forniscono agli sviluppatori componenti predefiniti, astrazioni e strumenti che snelliscono lo sviluppo di sistemi AI complessi.

Questi framework aiutano gli sviluppatori a concentrarsi sugli aspetti unici delle loro applicazioni fornendo approcci standardizzati alle sfide comuni nello sviluppo di agenti AI. Potenziano la scalabilità, l’accessibilità e l’efficienza nella costruzione di sistemi AI.

Introduction

This lesson will cover:

Learning goals

The goals of this lesson are to help you understand:

What are AI Agent Frameworks and what do they enable developers to do?

Traditional AI Frameworks can help you integrate AI into your apps and make these apps better in the following ways:

That all sounds great right, so why do we need the AI Agent Framework?

AI Agent frameworks represent something more than just AI frameworks. They are designed to enable the creation of intelligent agents that can interact with users, other agents, and the environment to achieve specific goals. These agents can exhibit autonomous behavior, make decisions, and adapt to changing conditions. Let’s look at some key capabilities enabled by AI Agent Frameworks:

So in summary, agents allow you to do more, to take automation to the next level, to create more intelligent systems that can adapt and learn from their environment.

How to quickly prototype, iterate, and improve the agent’s capabilities?

This is a fast-moving landscape, but there are some things that are common across most AI Agent Frameworks that can help you quickly prototype and iterate namely module components, collaborative tools, and real-time learning. Let’s dive into these:

Use Modular Components

SDKs like the Microsoft Agent Framework offer pre-built components such as AI connectors, tool definitions, and agent management.

How teams can use these: Teams can quickly assemble these components to create a functional prototype without starting from scratch, allowing for rapid experimentation and iteration.

How it works in practice: You can use a pre-built parser to extract information from user input, a memory module to store and retrieve data, and a prompt generator to interact with users, all without having to build these components from scratch.

Example code. Let’s look at an example of how you can use the Microsoft Agent Framework with AzureAIProjectAgentProvider to have the model respond to user input with tool calling:

# Esempio di Microsoft Agent Framework in Python

import asyncio
import os
from typing import Annotated

from agent_framework.azure import AzureAIProjectAgentProvider
from azure.identity import AzureCliCredential


# Definisci una funzione di esempio per prenotare viaggi
def book_flight(date: str, location: str) -> str:
    """Book travel given location and date."""
    return f"Travel was booked to {location} on {date}"


async def main():
    provider = AzureAIProjectAgentProvider(credential=AzureCliCredential())
    agent = await provider.create_agent(
        name="travel_agent",
        instructions="Help the user book travel. Use the book_flight tool when ready.",
        tools=[book_flight],
    )

    response = await agent.run("I'd like to go to New York on January 1, 2025")
    print(response)
    # Esempio di output: Il tuo volo per New York del 1 gennaio 2025 è stato prenotato con successo. Buon viaggio! ✈️🗽


if __name__ == "__main__":
    asyncio.run(main())

What you can see from this example is how you can leverage a pre-built parser to extract key information from user input, such as the origin, destination, and date of a flight booking request. This modular approach allows you to focus on the high-level logic.

Leverage Collaborative Tools

Frameworks like the Microsoft Agent Framework facilitate the creation of multiple agents that can work together.

How teams can use these: Teams can design agents with specific roles and tasks, enabling them to test and refine collaborative workflows and improve overall system efficiency.

How it works in practice: You can create a team of agents where each agent has a specialized function, such as data retrieval, analysis, or decision-making. These agents can communicate and share information to achieve a common goal, such as answering a user query or completing a task.

Example code (Microsoft Agent Framework):

# Creazione di più agenti che lavorano insieme utilizzando il Microsoft Agent Framework

import os
from agent_framework.azure import AzureAIProjectAgentProvider
from azure.identity import AzureCliCredential

provider = AzureAIProjectAgentProvider(credential=AzureCliCredential())

# Agente di recupero dati
agent_retrieve = await provider.create_agent(
    name="dataretrieval",
    instructions="Retrieve relevant data using available tools.",
    tools=[retrieve_tool],
)

# Agente di analisi dei dati
agent_analyze = await provider.create_agent(
    name="dataanalysis",
    instructions="Analyze the retrieved data and provide insights.",
    tools=[analyze_tool],
)

# Eseguire gli agenti in sequenza su un compito
retrieval_result = await agent_retrieve.run("Retrieve sales data for Q4")
analysis_result = await agent_analyze.run(f"Analyze this data: {retrieval_result}")
print(analysis_result)

What you see in the previous code is how you can create a task that involves multiple agents working together to analyze data. Each agent performs a specific function, and the task is executed by coordinating the agents to achieve the desired outcome. By creating dedicated agents with specialized roles, you can improve task efficiency and performance.

Learn in Real-Time

Advanced frameworks provide capabilities for real-time context understanding and adaptation.

How teams can use these: Teams can implement feedback loops where agents learn from interactions and adjust their behavior dynamically, leading to continuous improvement and refinement of capabilities.

How it works in practice: Agents can analyze user feedback, environmental data, and task outcomes to update their knowledge base, adjust decision-making algorithms, and improve performance over time. This iterative learning process enables agents to adapt to changing conditions and user preferences, enhancing overall system effectiveness.

What are the differences between the Microsoft Agent Framework and Azure AI Agent Service?

There are many ways to compare these approaches, but let’s look at some key differences in terms of their design, capabilities, and target use cases:

Microsoft Agent Framework (MAF)

The Microsoft Agent Framework provides a streamlined SDK for building AI agents using AzureAIProjectAgentProvider. It enables developers to create agents that leverage Azure OpenAI models with built-in tool calling, conversation management, and enterprise-grade security through Azure identity.

Use Cases: Building production-ready AI agents with tool use, multi-step workflows, and enterprise integration scenarios.

Here are some important core concepts of the Microsoft Agent Framework:

Azure AI Agent Service

Azure AI Agent Service is a more recent addition, introduced at Microsoft Ignite 2024. It allows for the development and deployment of AI agents with more flexible models, such as directly calling open-source LLMs like Llama 3, Mistral, and Cohere.

Azure AI Agent Service provides stronger enterprise security mechanisms and data storage methods, making it suitable for enterprise applications.

It works out-of-the-box with the Microsoft Agent Framework for building and deploying agents.

This service is currently in Public Preview and supports Python and C# for building agents.

Using the Azure AI Agent Service Python SDK, we can create an agent with a user-defined tool:

import asyncio
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient

# Definire le funzioni dello strumento
def get_specials() -> str:
    """Provides a list of specials from the menu."""
    return """
    Special Soup: Clam Chowder
    Special Salad: Cobb Salad
    Special Drink: Chai Tea
    """

def get_item_price(menu_item: str) -> str:
    """Provides the price of the requested menu item."""
    return "$9.99"


async def main() -> None:
    credential = DefaultAzureCredential()
    project_client = AIProjectClient.from_connection_string(
        credential=credential,
        conn_str="your-connection-string",
    )

    agent = project_client.agents.create_agent(
        model="gpt-4o-mini",
        name="Host",
        instructions="Answer questions about the menu.",
        tools=[get_specials, get_item_price],
    )

    thread = project_client.agents.create_thread()

    user_inputs = [
        "Hello",
        "What is the special soup?",
        "How much does that cost?",
        "Thank you",
    ]

    for user_input in user_inputs:
        print(f"# User: '{user_input}'")
        message = project_client.agents.create_message(
            thread_id=thread.id,
            role="user",
            content=user_input,
        )
        run = project_client.agents.create_and_process_run(
            thread_id=thread.id, agent_id=agent.id
        )
        messages = project_client.agents.list_messages(thread_id=thread.id)
        print(f"# Agent: {messages.data[0].content[0].text.value}")


if __name__ == "__main__":
    asyncio.run(main())

Core concepts

Azure AI Agent Service has the following core concepts:

Use Cases: Azure AI Agent Service is designed for enterprise applications that require secure, scalable, and flexible AI agent deployment.

What’s the difference between these approaches?

It does sound like there is overlap, but there are some key differences in terms of their design, capabilities, and target use cases:

Still not sure which one to choose?

Use Cases

Let’s see if we can help you by going through some common use cases:

Q: I’m building production AI agent applications and want to get started quickly

A: The Microsoft Agent Framework is a great choice. It provides a simple, Pythonic API via AzureAIProjectAgentProvider that lets you define agents with tools and instructions in just a few lines of code.

Q: I need enterprise-grade deployment with Azure integrations like Search and code execution

A: Azure AI Agent Service is the best fit. It’s a platform service that provides built-in capabilities for multiple models, Azure AI Search, Bing Search and Azure Functions. It makes it easy to build your agents in the Foundry Portal and deploy them at scale.

Q: I’m still confused, just give me one option

A: Start with the Microsoft Agent Framework to build your agents, and then use Azure AI Agent Service when you need to deploy and scale them in production. This approach lets you iterate quickly on your agent logic while having a clear path to enterprise deployment.

Let’s summarize the key differences in a table:

Framework Focus Core Concepts Use Cases
Microsoft Agent Framework Streamlined agent SDK with tool calling Agents, Tools, Azure Identity Building AI agents, tool use, multi-step workflows
Azure AI Agent Service Flexible models, enterprise security, Code generation, Tool calling Modularity, Collaboration, Process Orchestration Secure, scalable, and flexible AI agent deployment

Can I integrate my existing Azure ecosystem tools directly, or do I need standalone solutions?

La risposta è sì: puoi integrare i tuoi strumenti dell’ecosistema Azure esistenti direttamente con Azure AI Agent Service, soprattutto perché è stato progettato per funzionare perfettamente con gli altri servizi Azure. Puoi ad esempio integrare Bing, Azure AI Search e Azure Functions. C’è anche una profonda integrazione con Microsoft Foundry.

Il Microsoft Agent Framework si integra inoltre con i servizi Azure tramite AzureAIProjectAgentProvider e l’identità Azure, permettendoti di chiamare i servizi Azure direttamente dagli strumenti del tuo agente.

Esempi di codice

Hai altre domande sui framework per agenti AI?

Unisciti al Microsoft Foundry Discord per incontrare altri studenti, partecipare alle office hours e ottenere risposte alle tue domande sugli agenti AI.

Riferimenti

Lezione precedente

Introduzione agli agenti AI e casi d’uso

Lezione successiva

Comprendere i pattern di design agentico

Dichiarazione di non responsabilità: Questo documento è stato tradotto utilizzando il servizio di traduzione AI Co-op Translator. Pur impegnandoci per la correttezza, si prega di notare che le traduzioni automatiche possono contenere errori o inesattezze. Il documento originale nella sua lingua originale deve essere considerato la fonte autorevole. Per informazioni critiche, si raccomanda una traduzione professionale effettuata da un traduttore umano. Non ci assumiamo alcuna responsabilità per eventuali incomprensioni o interpretazioni errate derivanti dall’uso di questa traduzione.

This site is open source. Improve this page.