# Coding agents require skilled operators

<https://simonwillison.net/2025/Jun/18/coding-agents/#atom-everything>

- Host: [simonwillison.net](https://links.thingelstad.xyz/hosts/simonwillison.net/)
- Saved: June 19, 2025 (2025-06-19T17:12:23Z)
- Record: https://links.thingelstad.xyz/links/8b405227b581/
- JSON: https://links.thingelstad.xyz/links/8b405227b581.json

## What Jamie wrote

The narrative of having LLMs displace software developers needs to go away. The two are not in opposition, but instead are in an act or co-creation and enabling and supporting human developers to operate at a scale and level of understanding that is not possible otherwise. 

My entire career has had one very consistent fact:

**Our demand for software seems to have no bound**. 

I often say we have 20 lbs of flour and a 5 lb bag. It's been like that for decades. Leading software teams is often critically about what you are not going to do since it is entirely impossible to do everything. If we have huge impact on productivity -- I’m very positive we will simply want more.

**There is so much to build!**

There is another angle that isn't discussed as often. As software and systems have gotten more complex the challenge for engineers has gotten so much harder. It is common for developers to only know their little bit of code and nothing about the rest. This is dangerous but necessary. One brain simply cannot hold all of that context. Remember that when you load a product detail page on Amazon over 1,000 services are engaged to show that. Likely nobody actually knows how that is all related. 

AI in our development environments may help scale the ability for developers to actually see and understand these systems. To increase the confidence in what we are creating and modifying.

## Facts

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

## Appears in

- [WT324: Weekly Thing 324 / Agents, Shortcuts, Joy](https://weekly.thingelstad.com/archive/324/), Notable

## Same host

- [Highlights from the Claude 4 system prompt](https://links.thingelstad.xyz/links/8ff0eb636351/) (simonwillison.net, May 29, 2025)
- [Large Language Models can run tools in your terminal with LLM 0.26](https://links.thingelstad.xyz/links/17c8a3bb1985/) (simonwillison.net, May 29, 2025)
- [AI-assisted development needs automated tests](https://links.thingelstad.xyz/links/93b2d71dfa81/) (simonwillison.net, May 29, 2025)
- [Design Patterns for Securing LLM Agents against Prompt Injections](https://links.thingelstad.xyz/links/161cfcb03e8b/) (simonwillison.net, June 16, 2025)
- [Agentic Coding: The Future of Software Development with Agents](https://links.thingelstad.xyz/links/cc95c57c26c7/) (simonwillison.net, July 6, 2025)

## Saved the same week

- [Is there a Half-Life for the Success Rates of AI Agents? — Toby Ord](https://links.thingelstad.xyz/links/fd363642e7c1/) (tobyord.com, June 19, 2025)
- [Building Effective AI Agents Anthropic](https://links.thingelstad.xyz/links/560cdb7361bd/) (anthropic.com, June 19, 2025)
- [Websites Are Tracking You Via Browser Fingerprinting | Texas A&M University Engineering](https://links.thingelstad.xyz/links/b777e4d3226b/) (engineering.tamu.edu, June 19, 2025)
- [Yancey Strickler: Forget hustle culture. Behold the Artist Corporation | TED Talk](https://links.thingelstad.xyz/links/981c54ceb3fc/) (ted.com, June 19, 2025)
- [The Conversational SDLC | Joseph Bironas | Medium](https://links.thingelstad.xyz/links/68941166ba38/) (josebiro.medium.com, June 19, 2025)

## Elsewhere

- Wayback Machine: https://web.archive.org/web/*/https://simonwillison.net/2025/Jun/18/coding-agents/#atom-everything