All courses
Hamburger Icon

Foundations of Agentic AI

Course Summary

The Foundations of Agentic AI course focuses on building real-world chatbots and autonomous problem-solving agents using open-source tools, including Hugging Face Transformers, LlamaIndex, and OpenAI Gym. Participants will gain hands-on experience designing, building, and deploying intelligent agents capable of reasoning and interacting in natural language, powered by Hugging Face’s free pre-trained models. An optional 3rd-day hackathon allows teams to apply their knowledge to tackle realistic AI challenges using real-world datasets and APIs.

Prerequisites
In order to succeed in this course, you will need:

  • A basic understanding of machine learning and natural language processing
  • Proficiency in Python programming

Purpose
Gain experience designing, building, and deploying intelligent agents capable of reasoning and interacting in natural language
Audience
Those looking to design, build, and deploy conversational and problem-solving agents
Role
Software Developers | Data Scientists | Data Engineers | AI/ML Developers
Skill level
Intermediate
Style
Lecture | Hands-on Activities | Labs
Duration
2 days (3 days with optional hackathon)
Related technologies
AI/ML | Gen AI | Hugging Face | Python

 

Course objectives
  • Understand how to leverage Hugging Face models for building conversational and problem-solving agents
  • Build practical agentic AI systems using Python-based tools and frameworks
  • Integrate free and open-source libraries like LlamaIndex and OpenAI Gym for agent design
  • Deploy fully functional chatbots and agents without reliance on paid APIs

What you'll learn:

In this Foundations of Agentic AI course, you'll learn:

Foundations of Chatbots and Intelligent Agents

  • Introduction to Agentic AI and Hugging Face
    • What is Agentic AI?
      • Overview of agents, autonomy, and decision-making
    • Hugging Face Ecosystem
      • Introduction to Transformers, pipelines, and pre-trained models
    • Exploring a pre-built chatbot using Hugging Face's models
  • Building Chatbots with Hugging Face Transformers
    • Transformers for NLP Tasks: Understanding how Hugging Face’s models handle tasks like text generation Q&A, and summarization
      • Loading a pre-trained conversational model (e.g., DialoGPT or BlenderBot)
      • Fine-tuning the chatbot on a custom dataset (e.g., customer service FAQs)
      • Testing the chatbot’s responses and improve performance
  • Augmenting Chatbots with LlamaIndex
    • Connecting Chatbots to External Data
      • Making chatbots more intelligent with LlamaIndex
      • Retrieval-Augmented Generation (RAG)
      • Answering queries with real-time information
    • Building a chatbot knowledge base (e.g., FAQs, company policies)
      • Testing the chatbot on domain-specific questions
  • Introduction to Simulated Environments with OpenAI Gym
    • Simulating Tasks for Intelligent Agents
      • Why simulations are essential for agent training
    • Customizing OpenAI Gym Environments
      • Designing environments for specific tasks
    • Setting up a basic OpenAI Gym environment (e.g., Taxi-v3)
      • Building a rule-based agent to solve a simple navigation task

Advanced Agent Design and Real-World Applications

  • Multi-Modal Agents and Advanced Integrations
    • Multi-Modal Inputs for Agents
      • Combining text, images, and structured data
    • Integrating Hugging Face models for text and vision tasks (e.g., image captioning with CLIP or BLIP)
      • Creating a multi-modal chatbot that processes both text and images
  • Problem-Solving Agents with OpenAI Gym
    • Reinforcement Learning Primer: Basics of Q-learning and policy optimization
    • Training a reinforcement learning agent in a custom Gym environment (e.g., resource allocation or delivery)
      • Comparing a rule-based agent’s performance with an RL-based agent
  • Deploying Agentic AI Systems
    • Deploying Chatbots and Agents Locally or on Cloud Platforms
      • Tools and best practices
    • Deploying a chatbot locally using Flask or FastAPI
      • Build a simple front-end to interact with the deployed agent

Foundations of Agentic AI

real world content

Real-World Content

Project-focused demos and labs using your tool stack and environment, not some canned "training room" lab.
expert practitioners

Expert Practitioners

Industry experts that bring their battle scars into the classroom.
experiential learning

Experiential Learning

More coding than lecture, coupled with architectural and design discussions.
fully customized

Tailored Outlines

Once-size-fits-all doesn't apply to training teams. That's where we come in!

Dive in and learn more

When transforming your workforce, it’s important to have expert advice and tailored solutions. We can help. Tell us your unique needs and we'll explore ways to address them.

Let's chat

By filling out this form and clicking submit, you acknowledge our privacy policy.