The primary tools you are going to be using for this course are Visual Studio Code, a local Python environment, and a large language model that I will provide for you to use.
If you have not done so already, you should download and install Visual Studio Code from https://code.visualstudio.com/download.
If you have not already set up homebrew on your Mac, follow the instructions to install Homebrew at https://brew.sh/
Next, run these commands in a terminal:
brew install --cask miniforge
conda init "$(basename "${SHELL}")"
mkdir work
cd work
conda create -n agents
conda activate agents
conda install jupyter
conda install langchain langchain-openai
These commands will set up a work directory in your home directory, and then set up a python environment named "agents" inside that directory. The install commands will install the initial set of packages you will need in that Python environment.
Start by downloading the windows installer for Miniforge from https://conda-forge.org/download/
After you run the installer, go to the start menu and type miniforge to access the miniforge prompt application. In that prompt application type these commands:
mkdir work cd work conda create -n agents conda activate agents conda install jupyter conda install langchain langchain-openai
These commands will set up a work directory in your home directory, and then set up a python environment named "agents" inside that directory. The install commands will install the initial set of packages you will need in that Python environment.
Click this link to download the first example notebook we will be working with. After downloading, drag that file from your downloads folder into the work folder in your home folder.
Start up Visual Studio Code and open the work directory in your home directory. Then double-click the 01-langchain_examples.ipynb file to open it.
In the upper right hand corner of the window for the notebook you should see a Select Kernel button. Click that button and select the agents environment you set up in the previous step.
Bug alert! Visual Studio Code uses the Jupyter extension to enable you to work with Jupyter notebooks. Recent versions of that extension suffered from a bug that would cause the Select Kernel button to not work correctly. If this happens to you, go to the extensions tab and look for the Jupyter extension. Click the gear icon in the lower right corner of the extension and select the Install Specific Version... command. That will pop up a list of versions to choose from - select version 2025.7.0 from that list.