Part One: Google Vertex AI
Objective: This lab will provide you with hands-on experience using a major cloud AI platform, Google's Vertex AI. You will learn how to activate cloud services, explore a professional AI development environment, and understand core concepts like tokens, prompt variables, and API keys that are essential for deploying AI solutions.
Connection to the Playbook: This lab provides a practical foundation for the concepts discussed in Chapter 7: This is my Rubik's Cube. You will be directly engaging with the Infrastructure and Model sides of the DIMASI cube, specifically exploring the "Buy" path via a Platform as a Service (PaaS) and API-as-a-Service model.
Step 0: Quick Approach
- If you have activated your trial of Gemini, visit https://gemini.google.com/ to explore the consumer facing version.
- Then visit aistudio.google.com to get the version for developers and enthusiasts.
Step 1: Activate Google Cloud Credits [The Full Back-end Access]
Your first step is to set up your Google Cloud account and apply the educational credits provided for the course. This will give you a budget to experiment with the platform's powerful tools without incurring personal costs.
- Find Your Link: Locate the unique link for activating your Google Cloud credits on the course Canvas page.
- You must REQUEST the credits using your university email address.
- Redeem with Your Personal Gmail Account:
- Crucial Warning: You must REDEEM these credits using a personal Gmail account (e.g.,
[email protected]).
- DO NOT use your university-provided email account (e.g.,
[email protected]). Any credits accidentally applied to your university account will be unusable for this course, as that account is managed by the university's central IT and has different permissions.
- Follow the On-Screen Instructions: Google will guide you through the process of creating a billing account (if you don't have one) and applying the credits. You may need to enter a credit card for verification, but you will not be charged as long as you stay within your free credit limit.
Step 2: First Look at Google Vertex AI
With your account activated, it's time to visit the central hub for Google's AI services.
- Navigate to Vertex AI: Go to the Google Cloud Console and use the search bar at the top to search for "Vertex AI". Click the result to go to the main Vertex AI page.
- Explore the Dashboard: You will land on the Vertex AI Dashboard. Take a moment to look around. This is your command center. You'll see options for managing datasets, training models, running batch predictions, and more. This dashboard gives you a sense of the full MLOps lifecycle discussed in Chapter 9: Duct Tape that Floats, from data preparation to deployment and monitoring. Don't worry about understanding everything; the goal here is to get a feel for the landscape.
Step 3: Experiment in the Vertex AI Studio
The Vertex AI Studio is an interactive "playground" where you can work directly with Google's powerful generative AI models.
- From the Vertex AI navigation menu on the left, find the "Generative AI Studio" section (it may be called "Vertex AI Studio") and click on Language.
- You are now in a prompt design studio. Here you can experiment with different models (like Gemini) and see how they respond to your instructions.
3.1: Understanding Tokens