Setting Up

Choose your platform below and follow the setup instructions. Each tab walks you through creating an account (for cloud platforms) or installing the software (for local platforms), so you are ready to start building AI agents.

Tip

You only need to set up one platform to follow this course. If you are unsure which to choose, see Choosing a Platform.

Claude

Claude is Anthropic’s AI assistant. Setting up takes about two minutes.

1

Go to claude.ai

Open your web browser and navigate to claude.ai. This is the official website for Claude.

2

Create your account

Click Sign Up. You can create an account using your email address, or sign up with your Google or Apple account for a faster setup. Enter your information and verify your email if prompted.

3

Complete the onboarding

After signing up, Claude may ask you a few quick questions about how you plan to use it. Answer these honestly; they help Claude tailor its responses to your needs. You can skip this step if the option is available.

4

Explore the interface

Once you are signed in, you will see the main chat interface. Here is what to look for:

• Message box: The text field at the bottom where you type your messages. This is where you will write your prompts.
• Conversation history: The left sidebar shows your past conversations. Each conversation is separate and has its own context.
• Model selector: Near the top or in the message area, you may see an option to choose which Claude model to use. The free tier gives you access to the default model.
• Projects: In the left sidebar, look for the Projects section. This is where you will later create persistent agents with custom instructions and uploaded knowledge files.

5

Send your first message

Type a simple message like “Hello! Can you tell me what you are?” and press Enter (or click the send button). Claude will respond, confirming that everything is working. Congratulations, you are ready to start the course!

Note

The free tier of Claude limits the number of messages you can send per day. If you hit the limit, you can wait for it to reset (usually a few hours) or upgrade to Claude Pro for higher limits.

ChatGPT

ChatGPT is OpenAI’s AI assistant. Setting up takes about two minutes.

1

Go to chatgpt.com

Open your web browser and navigate to chatgpt.com. This is the official website for ChatGPT.

2

Create your account

Click Sign up. You can register with your email address, or use your Google, Microsoft, or Apple account. Follow the prompts to set your name and verify your account.

3

Choose your plan

After signing up, you may be asked to choose a plan. Select Free to continue without payment. You can always upgrade later if you want access to more powerful models or additional features.

4

Explore the interface

Once signed in, you will see the ChatGPT interface. Here is what to look for:

• Message box: The text field at the bottom where you type your prompts.
• Conversation sidebar: The left panel shows your previous chats. Click any conversation to continue it.
• Model selector: At the top of the chat, you can choose which model to use. The free tier gives you access to the current standard model with daily usage limits.
• Explore GPTs: In the sidebar, you will find a section for GPTs. You can browse and use existing GPTs from the GPT Store. Creating your own Custom GPTs requires a paid plan (Plus, Team, or Enterprise).

5

Send your first message

Type “Hello! What can you help me with?” and press Enter. ChatGPT will respond with an overview of its capabilities. You are all set!

Note

ChatGPT’s free tier provides generous daily usage with the standard model. If you need higher limits or access to the latest models, ChatGPT Plus costs $20/month.

Gemini

Gemini is Google’s AI assistant. If you have a Google account, you can start immediately.

1

Go to gemini.google.com

Open your web browser and navigate to gemini.google.com.

2

Sign in with your Google account

Click Sign in and use your existing Google account (the same one you use for Gmail, YouTube, or Google Drive). If you do not have a Google account, click Create account and follow the registration process.

3

Accept the terms

On your first visit, Google may ask you to accept the terms of service for Gemini. Review them and click I agree or Continue to proceed.

4

Explore the interface

Once signed in, you will see the Gemini chat interface. Here is what to look for:

• Message box: The text input at the bottom of the screen where you type your prompts.
• Recent conversations: The left sidebar lists your previous chats with Gemini.
• Google integration: Gemini can connect to your Google services. Look for options related to Gmail, Drive, and Docs in the settings or through extensions.
• Gems: In the sidebar, you may see a Gems section. These are Gemini’s version of custom agents. Creating Gems requires Gemini Advanced (paid plan), but you can always paste instructions at the start of a conversation instead.

5

Send your first message

Type “Hi! What are you and what can you do?” and press Enter. Gemini will introduce itself and list its capabilities. You are ready to begin!

Tip

If you already use Google Workspace (Gmail, Docs, Drive), Gemini has a major advantage: it can access and work with your existing documents and emails directly. This will be especially useful in the projects section of the course.

Ollama

Ollama lets you run AI models locally on your computer. Setup takes about 5-10 minutes depending on your internet speed (you need to download the software and a model).

Caution

Ollama requires a reasonably modern computer. For a good experience, you should have at least 8 GB of RAM. A dedicated GPU (graphics card) is not required but will make responses significantly faster. If your computer is more than 5-6 years old, consider starting with a cloud platform instead.

1

Download Ollama

Go to ollama.com and click the Download button. Choose the installer for your operating system:

• Windows: Download the .exe installer
• Mac: Download the .dmg file
• Linux: Follow the one-line install command shown on the website

2

Install Ollama

Run the installer you downloaded:

• Windows: Double-click the .exe file and follow the installation wizard. Accept the default settings.
• Mac: Open the .dmg file and drag Ollama to your Applications folder.
• Linux: The install command from Step 1 handles everything automatically.

3

Start Ollama

After installation, launch Ollama:

• Windows: Ollama runs as a background service automatically after installation. You will see a small llama icon in your system tray (bottom-right corner of the screen).
• Mac: Open Ollama from your Applications folder. You will see a llama icon in your menu bar.
• Linux: Ollama starts automatically as a system service.

4

Download your first model

Open a terminal (Command Prompt on Windows, Terminal on Mac/Linux) and type:

ollama pull llama3.2

This downloads the Llama 3.2 model (about 2 GB). Wait for the download to complete; it may take a few minutes depending on your internet connection. Llama 3.2 is a great starting model because it is small, fast, and capable.

5

Start chatting

Once the model is downloaded, start a conversation by typing:

ollama run llama3.2

You will see a prompt where you can type messages directly. Type “Hello! Who are you?” and press Enter. The model will respond. To exit the chat, type /bye and press Enter.

Tip

You can see all available models by visiting ollama.com/library. Popular choices for beginners include llama3.2 (small and fast), mistral (good all-around), and gemma2 (Google’s open model). Download additional models anytime with ollama pull model-name.

LM Studio

LM Studio gives you a graphical desktop application for running AI models locally. Setup takes about 5-10 minutes.

Caution

Like Ollama, LM Studio runs models on your computer. You should have at least 8 GB of RAM for a reasonable experience. A dedicated GPU is helpful but not strictly required. If your computer struggles, start with a cloud platform.

1

Download LM Studio

Go to lmstudio.ai and click the download button for your operating system (Windows, Mac, or Linux). The application is free for personal use.

2

Install LM Studio

Run the installer:

• Windows: Double-click the downloaded .exe file and follow the installation prompts. Accept the default settings.
• Mac: Open the .dmg file and drag LM Studio to your Applications folder.
• Linux: Follow the instructions on the download page for your distribution.

3

Launch the application

Open LM Studio from your Applications folder (Mac), Start menu (Windows), or application launcher (Linux). The first time you open it, you may see a brief setup screen. Follow any prompts to complete the initial configuration.

4

Download your first model

Inside LM Studio, navigate to the Discover tab (it may also be labeled Search or have a magnifying glass icon). In the search bar, type Llama 3.2 or Gemma 2. You will see a list of available models.

Click the Download button next to a model. Look for one labeled with a size you are comfortable with. Models labeled “1B” or “3B” are smaller and will run on most computers. The download may take a few minutes.

5

Start chatting

Once the model finishes downloading, go to the Chat tab (usually at the top or in the left sidebar). Select the model you just downloaded from the model dropdown at the top of the chat area. Type “Hello! Tell me about yourself.” in the message box and press Enter.

LM Studio will process your message and display a response. If the response comes back, everything is working and you are ready to begin the course!

Tip

LM Studio’s interface feels very similar to ChatGPT, which makes it comfortable for beginners. You can adjust settings like temperature (creativity level) and context length using the sliders in the right panel. Do not worry about these settings for now. We will cover them in the Prompts & Configuration section.

Verifying Your Setup

No matter which platform you chose, you should now be able to:

1. Open the platform. Either in your web browser (Claude, ChatGPT, Gemini) or as a desktop application / terminal (Ollama, LM Studio).
2. Type a message. Write something in the message input area.
3. Get a response. The AI should reply within a few seconds (cloud) or a few seconds to a minute (local, depending on your hardware).

If any of these steps did not work, here are common fixes:

• Cloud platforms: Make sure you verified your email during signup. Try clearing your browser cache or using a different browser.
• Ollama: Make sure the Ollama service is running (check for the llama icon in your system tray or menu bar). If ollama run gives an error, try ollama serve in a separate terminal window first.
• LM Studio: Make sure you have actually downloaded a model (the chat will not work without one). If the application is very slow, try a smaller model (1B parameters instead of 7B).

Tip

You are now set up and ready to learn. The next section, Foundations, will teach you what AI agents actually are and how they work. Let’s go!