A free course on how Agents work by practically showing how to build your own using the latest libraries and tools. If you are bogged down by repetitive tasks, this course is certainly worth your time.
Agents is the new trend in the AI world. The concept is about delegating tasks to your bots to do a job autonomously like using digital tools or executing complex tasks on your computer. One simplistic way of looking at it, it's like having Autohotkey on steroids.
Autohotkey, of course is an open-source scripting language, that allows users to easily create small to complex scripts for all kinds of tasks such as: form fillers, auto-clicking, macros, etc. Take scripting language out and replace it with Agents which have much more capability; scrapping web sites, summarize articles and more.
This course then will show you how you can build your own agents to assign them any kind of tasks. You will:
Study AI Agents in theory, design, and practice.
Learn to use established AI Agent libraries such as smolagents, LlamaIndex, and LangGraph.
Share your agents on the Hugging Face Hub and explore agents created by the community.
Participate in challenges where you will evaluate your agents against other students’.
Optionally earn a certificate of completion by completing assignments (until May 1st).
Unit 1. Introduction to Agents What an Agent is and more generally what a LLMs are. Understanding AI Agents through the Thought-Action-Observation Cycle and enabling the Agent to engage with Its environment. Create a first simple Agent using the smolagents framework.
Unit 2. Frameworks for AI Agents The available frameworks for building AI Agents with more detailed look of the smolagents framework. Building Agents that use code and writing actions as code snippets and then generate Python tool calls to perform actions, achieving action representations that are efficient, expressive, and accurate.
Unit 2.1 The smolagents framework
What kind of agents can there be build:
Tool Calling Agents
Retrieval Agents
Multi-Agent Systems
Vision and Browser agents
and creating proper smallagents.
Unit 2.2 The LlamaIndex framework Introduction to LLamaIndex, what are Components in LlamaIndex, using Agents and creating Agentic workflows in LlamaIndex.
Bonus Unit 1. Fine-tuning an LLM for Function-calling What is Function Calling and fine-tuning your model for Function-calling
The prerequisites for taking the course are pretty light on demands:
Basic knowledge of Python
Basic knowledge of LLMs
A free Hugging Face account to push and load models, agents, and create Spaces
All in all, this is the perfect beginner's course to get into the world of Agentic AI and learn how to automate recurring, mundane tasks to free you and let you focus on the creative parts you enjoy most.
We are all being urged to work in Rust because its safer, but this is easier said than done. The embedded world really needs a memory-safe language, but Rust gets surprisingly little support from manu [ ... ]
The Java SDK implementation of the Model Context Protocol gets to pre-release version 0.70, opening the way for Java to enjoy transparent interoperability with AI services.
