Question 1

Tool calling vs function calling vs structured output — what changed in 2025?

Accepted Answer

**Target Audience:** Engineers who keep meeting three overlapping names in docs, SDKs and interviews and want one clear mental model plus an accurate timeline.
**Difficulty:** Medium
**Verified with:** Node 22.22.0 and Python 3.12.3 (the wire-format adapter below, run against canned API responses). Dates and API shapes are cited inline with their source, checked October 2026. The title says 2025; this answer covers 2025 and notes what has changed since.

> **How to read this doc:** Every idea is

Question 2

How do you design good tool definitions for an LLM?

Accepted Answer

**Target Audience:** Engineers turning a backend (REST endpoints, database queries, internal scripts) into tools for an LLM or agent, and anyone who has watched a model pick the wrong tool.
**Difficulty:** Medium
**Verified with:** Node 22.22.0 and Python 3.12.3 (a tool-definition linter and a tool-selection eval). Vendor guidance is quoted from the Claude docs ("Define tools", "Handle tool calls"), Anthropic engineering posts and other sources as cited, checked October 2026.

> **How to read th

Question 3

How does function calling (tool use) work under the hood?

Accepted Answer

**Target Audience:** Engineers who have passed a `tools` array to an LLM API and want to know exactly what happens between "the model decided to call a tool" and "the answer came back".
**Difficulty:** Medium
**Verified with:** Node 22.22.0 and Python 3.12.3 (a full tool-use loop with a stub model that reads and writes plain text). API behaviour is quoted from the Claude docs ("Define tools", "Handle tool calls", "Stop reasons"), checked October 2026, and other vendors where cited.

> **How to r

AI ENGINEERINGDECODED.

Medium

Tool calling vs function calling vs structured output — what changed in 2025?

How do you design good tool definitions for an LLM?

How does function calling (tool use) work under the hood?