Building with the Claude API
Every chatbot, agent, and LLM feature you ship eventually comes down to one HTTP call. Building with the Claude API teaches that call and everything downstream of it: the one idea that explains the rest, and the production concerns most tutorials never reach.
Across eleven parts it covers the request and response model, how conversation memory really works, streaming, the tool-use loop you actually have to write, structured JSON output, prompt caching, cost estimation before you send, the batch API, rate limits and 429s, and the logging that turns a black box into something you can operate.
Start at Part 1, or jump to any part below.
The parts
- Your First Claude API Call, and the One Idea That Explains Everything Downstream
- Your Chatbot's Memory Is a Lie. Here Is What Is Really Going On.
- Claude API Streaming: The Event Protocol Behind the Typing Cursor
- Function Calling With Claude Is Not What You Think: Here Is the Loop You Actually Have To Write
- Stop Parsing the Model's Prose: How to Get Real JSON Out of the Claude API
- Your Claude API Bill Is Mostly Wasted. One Parameter Cuts It by Ninety Percent.
- The Claude API Endpoint That Tells You What a Request Will Cost Before You Send It
- Your Overnight LLM Backlog Is Costing You Double. Here Is How to Stop Paying It.
- Your Claude API Integration Is Running With Most of Its Features Switched Off
- Your Claude API Demo Has Never Seen a 429. Production Will.
- Your Claude Integration Is a Black Box Until You Log This One Header