Build AI Apps with Python: Currency Converter with Function Calling | Episode 7

Claude can now DO things! In this episode, we teach Claude to call a Python function — a currency converter that converts USD to Singapore dollars, euros, yen, and more. This is the start of Phase 2: Tool Use & Function Calling. You define a function, describe it with a JSON schema, and Claude decides when to call it. The tool use loop: user asks → Claude calls tool → you execute → Claude answers naturally. Student code: https://github.com/GoCelesteAI/build-ai-apps-python/tree/main/episode07 Every keystroke is shown on screen with 3-second pauses so you can follow along at your own pace. What You'll Learn: • Defining tools with JSON schema (name, description, input_schema) • The tools parameter in messages.create() • Mock exchange rates dictionary • Building a convert_currency() function • Checking stop_reason for "tool_use" • Extracting tool_block with name and arguments • Executing the function with **tool_block.input • Sending tool_result back to Claude for a natural answer • The complete tool use loop • Running Python scripts with :!python % Timestamps: 0:00 - Introduction 0:12 - What is Tool Use? (Preview) 0:44 - Creating first_tool.py 1:00 - Imports and setup 2:10 - Mock exchange rates dictionary 3:52 - convert_currency() function 5:22 - Save progress 5:32 - Tool definition — the JSON schema 6:25 - Input schema with properties 7:40 - Save tool definition 8:12 - API call with tools parameter 9:42 - Checking stop_reason == "tool_use" 10:18 - Extracting tool_block 11:05 - Executing the function 11:38 - Sending tool_result back to Claude 13:18 - Tool result block with ID 14:38 - Claude's final answer print 15:48 - Save and run 16:08 - The complete tool use loop in action! 16:25 - Code review 16:45 - Recap: 3 Key Takeaways 17:17 - End Screen Key Takeaways: 1. Tools are JSON schemas that describe your functions — Claude reads them to decide what to call 2. Check stop_reason for "tool_use" — then execute the function and return the result 3. The full loop: ask → call → execute → answer — that is function calling