Bootcamp
Build Agentic Apps

Build Agentic Apps

Agents analyzing cryptocurrencies

Agents analyzing cryptocurrencies

Up until now, we've been working with individual AI models to perform specific tasks. But it turns out, AI models can work together in a team to accomplish more complex tasks. This is where the concept of Agents comes in.

Tutorial Goals

In this tutorial you will:

  • Learn what agents are and how to use them to complete complex tasks
  • Understand the basics of CrewAI - install and build simple app
  • Learn how to provide tools to your agents that they can use
  • Develop a team (crew) to do cryptocurrency market analysis

What is an Agent?

Intuitively, an agent is a software program that acts on behalf of a user or a system. In the context of AI, an agent is an entity that can perceive its environment, make decisions, and act upon the environment to achieve its goals. Specifically, we'll use LLMs to power the "brains" of these agents.

One of the most used Agent concept is ReAct1 which is a conceptual framework for building AI agents that can reason and act in a unified manner. The idea is to combine the reasoning capabilities of language models with the ability to perform actions in a given environment.

Agents with CrewAI

MLExpert is loading...

References

Footnotes

  1. ReAct: Synergizing Reasoning and Acting in Language Models (opens in a new tab)

  2. crewAI (opens in a new tab)

  3. CrewAI Tools (opens in a new tab)

  4. LangChain Tools (opens in a new tab)

  5. CrewAI Agents (opens in a new tab)