GCP-CGAIA: Create Generative AI Apps on Google Cloud

Module 1: Generative AI Applications

This module discusses the use of generative AI in applications, the foundation models provided by Google, and the challenges encountered when building generative AI applications.

Objectives:

• Describe generative-AI-based application types and use cases.
• Describe foundation models provided by Google.
• Discuss challenges of generative AI for applications.

Module 2: Prompts

This module discusses prompts for generative AI models and how to best design prompts to improve effectiveness of model responses.

Objectives:

• Describe how to build prompt templates to improve model response quality in applications.
• Describe the essential and optional components of a prompt.
• Describe how to use a prompt template in an application

Module 3: Get Started with Vertex AI Studio

In this module, you experiment with Vertex AI Studio, which provides tools to rapidly prototype, tune models with your own data, and seamlessly deploy to applications. You explore multimodal capabilities of Gemini, design prompts, and generate conversations.

Objectives:

Design prompts within Vertex AI Studio

Module 4: Get Started with Vertex AI Studio

This module discusses how foundation model accuracy can be improved with retrieval augmented generation (RAG) and how vector embeddings can be used to find matching data for a user’s query.

Objectives:

• Describe methods for improving foundation model accuracy.
• Describe how vector embeddings can be used to retrieve relevant RAG data for a given query.
• Describe the components of a RAG-capable generative AI application on Google Cloud.

Module 5: Build an LLM and RAG-based Application

In this module, you build a chat application that uses large language models (LLMs) and retrieval augmented generation (RAG) to create engaging and informative conversations.

Objectives:

• Build an LLM and RAG-based chat application.