# LLMs are functions, not brains.

<https://james-pritchard.com/blog/llms-are-functions>

- Host: [james-pritchard.com](https://links.thingelstad.xyz/hosts/james-pritchard.com/)
- Saved: May 20, 2026 (2026-05-20T12:31:41Z)
- Record: https://links.thingelstad.xyz/links/044e16db1b61/
- JSON: https://links.thingelstad.xyz/links/044e16db1b61.json

## What Jamie wrote

When building agentic systems a big part of what you design is the guardrails, the context management, and tool environment for the LLM to operate in, to "be" the agent you are defining. This article is a good overview of the structure and how you build around the LLM capabilities. The naive view would be that you just give some input to a prompt. Not at all, the harness and capabilities are the thing that truly make it powerful.

## Facts

- Kind: [Page](https://links.thingelstad.xyz/kinds/page/)

## Appears in

- [WT349: WT349 — Owning the Rails](https://weekly.thingelstad.com/archive/349/), Notable

## Saved the same week

- [A Field Guide to Learning - Bite of an Apple](https://links.thingelstad.xyz/links/3bce2da644f3/) (brianschrader.com, May 20, 2026)
- [feedr: A feature-rich terminal-based RSS/Atom feed reader written in Rust.](https://links.thingelstad.xyz/links/ad843f04e48c/) (github.com, May 20, 2026)
- [Project Glasswing: what Mythos showed us](https://links.thingelstad.xyz/links/9cce0b5a8441/) (blog.cloudflare.com, May 20, 2026)
- [How to deal with your kid leaving](https://links.thingelstad.xyz/links/3bb580e9cd29/) (buttondown.com, May 20, 2026)
- [zerolang: The programming language for agents](https://links.thingelstad.xyz/links/ff231836a24a/) (github.com, May 20, 2026)

## Elsewhere

- Wayback Machine: https://web.archive.org/web/*/https://james-pritchard.com/blog/llms-are-functions