Prompt Engineering with LangChain

Introduction

• Create powerful LLM driven applications

1. Introduction to Language Models

• What are language models?

2. LLMs and Text Generation

• How do language models generate text?
• Base LLMs vs. instruction-tuned LLMs
• Training, fine-tuning, and in-context learning
• Prompt engineering

3. Components of LangChain

• What is LangChain?
• LangChain overview
• Model I/O: Interface with language models
• Retrieval: Interface with application-specific data
• Chains: Construct sequences of calls
• Agents: Let chains choose tools based on high-level directives
• Memory: Persist application state between runs of a chain

4. Basics of Prompting

• Prompt basics
• Principles and tactics for prompting

5. Prompt Templates Deep Dive

• Introduction to prompt templates
• Multi-input prompt templates
• Chat prompt template
• Serializing prompts
• Zero-shot prompts
• Custom prompt templates
• Prompt pipelining
• Chat prompt pipelining
• Prompt composition
• Few-shot prompt templates
• Few-shot prompt templates for chat
• Introduction to example selectors
• Length-based example selector
• Max marginal relevance example selector
• N-gram overlap example selector
• Semantic similarity example selector
• Partial prompt templates

6. Prompting Techniques

• Chain of thought
• Self-consistency
• Self-ask
• ReAct
• RAG
• FLARE
• Plan and execute

7. Prompt Management a.k.a. PromptOps

• Prompt management
• LangSmith
• LangSmith walkthrough
• Prompt versioning in LangSmith
• LangSmith deep dive
• Managing prompt length for agents

8. The LLM Landscape

• Applications of language models
• The LLM landscape