Application Development with LLMs on Google Cloud
Contact us to book this course
Generative AI
On-Site, Virtual
1 day
In this course, you'll dive into the details of using Large Language Models (LLMs) in your applications. You'll start by exploring the core principles that underpin prompting LLMs, providing a foundational understanding of how they work.
Next, you will focus on Google's latest family of models, Gemini. You'll explore the various Gemini models (Flash-lite, Flash, Pro) and their multimodal capabilities, learning how to access and utilize them through the Google AI APIs and Vertex AI. This includes a deep dive into effective prompt design and engineering within the Vertex AI Studio environment.
Then, the course moves to application development frameworks. You'll first learn about LangChain, an open-source framework designed to simplify the creation of complex, data-aware, and agentic applications powered by language models. This section will cover chains, agents, memory, and retrieval augmented generation (RAG). You will then discuss similar concepts in other developer kits, like the Agent Development Kit (ADK) and LangChain, to provide a broader perspective on application development tools.
Course objectives
- Explore the different options available for using generative AI on Google Cloud
- Use Vertex AI Studio to test prompts for large language models
- Develop LLM-powered applications using generative AI
- Apply prompt engineering techniques to improve the output from LLMs
- Build a multi-turn chat application using the Gemini API and LangChain.
Audience
Application developers and others who wish to leverage LLMs in applications.
Prerequisites
Completion of Introduction to Developer Efficiency on Google Cloud or equivalent knowledge.
Course outline
- What is Generative AI?
- Vertex AI on Google Cloud
- Generative AI Options on Google Cloud
- Introduction to the Course Use Case
- Vertex AI Studio
- Designing and Testing Prompts
- Data Governance in Vertex AI Studio
- Lab: Getting Started with the Vertex AI Studio User Interface
- Introduction to Grounding
- Integrating the Vertex AI Gemini APIs
- Chat, Memory and Grounding
- Search Principles
- Lab: Getting Started with Gen AI + Vertex AI Gemini API
- Review of few-shot prompting
- Chain-of-thought prompting
- Meta Prompting, Multistep, and Panel Prompts
- RAG and ReAct
- Lab: Advanced Prompt Architectures
- LangChain for Chatbots
- ADK for Chatbots
- Chat Retrieval
- Lab: Implementing RAG Using LangChain
