Overview

This course teaches developers how to build applications with Claude's API. Through seven modules, you'll learn API fundamentals, prompt engineering, tool use, retrieval systems, and agentic workflows. By the end, you'll be able to build production applications that use Claude for tasks ranging from simple requests to autonomous agents. Learners will build applications including chat interfaces, evaluation pipelines, tool-enabled systems, retrieval augmented generation implementations, and autonomous agents. These projects apply skills in API integration, prompt engineering, tool design, vector search, and workflow orchestration.

Syllabus

Getting Starting with Claude

In this module, you'll learn the fundamentals of working with Claude's API to build AI-powered applications. You'll start by understanding the different Claude model variants and how to choose the right one for your use case. Then you'll learn to authenticate, make requests, and handle responses.

Prompt Engineering & Evaluation

In this module, you'll learn systematic approaches to creating and testing effective prompts for Claude. You'll start by building evaluation pipelines that let you test prompts objectively using automated grading methods. You'll learn to generate comprehensive test datasets that expose both strengths and weaknesses in your prompts.

Claude Features and Tool Use

In this module, you'll explore Claude's advanced features and learn to build applications where Claude can call functions and interact with external systems. You'll start with features like extended thinking for complex reasoning, image and PDF support for multimodal applications, and prompt caching to reduce latency and costs.

Model Context Protocol (MCP)

In this module, you'll learn the Model Context Protocol, a standardized way to build servers and clients that provide tools and resources to Claude applications. MCP lets you create reusable integrations that work across different applications instead of building custom tool implementations for each project.

Retrieval Augmented Generation (RAG)

In this module, you'll learn to build retrieval augmented generation systems that provide Claude with relevant information from your documents and data. RAG addresses Claude's context limits by retrieving only the relevant portions of large document collections based on user queries.

Claude Code & Computer Use

In this module, you'll learn to use Claude Code for development tasks and build applications using Computer Use for automated interface interactions. Claude Code accelerates development by helping with code generation, refactoring, debugging, and documentation tasks.

Agentic Workflows

In this module, you'll learn to design and implement agentic workflows that handle complex, multi-step tasks. You'll start by understanding when to use workflows versus simple prompts, then learn three core workflow patterns: parallelization for executing multiple tasks simultaneously, chaining for sequential processing where outputs feed into subsequent steps, and routing for conditional branching based on intermediate results.