AI Is Not a Dev

I’ve been in the industry for 30 years professionally and before that 10 years as a hobbyist assembly language (65C02, 68K, PPC and finally x86) programmer. To pretend that the previous “4GL” tools are analogous to what LLMs can do is silly.

I am always using the anecdote that there are projects I scoped pre-2023 that I would have had to have at least one or two junior devs to do help me do the work after I wrote out all of the design specs and now I can do myself in the same amount of time with ChatGPT. That’s along with my lead/architect role of talking to the client, a lot of project management work, design documents, helping sales etc.

It’s a step change.

While I hated every single 4GL tool that I’ve ever encountered (well HyperCard has a soft spot in my heart), because it was limiting, I can use ChatGPT, to create my infrastructure (CDK, Cloudformation, Terraform), and my code based on talking to it like I would a good junior developer.

I can’t tell a junior developer I need IAC done in Node (the CDK), and code written in Python using the SDK and expect them to immediately know how to use both across 130+ different services.

I have never once in my decade plus of being over projects and having the ear (or direct authority) to hire someone said “what I really need are some junior developers who don’t have any real world experience. That would definitely help me ship faster”.

Why would I do that instead of poaching a mid level developer who can hit the ground running without doing a year of negative work (ie take time away from the team asking dumb questions).

This hammers starts sounding like Excalibur given all the mythos around it. :)

Agreed on the minimal-chargeable-bar point. The beauty arises when juniors grow.

It’s not a human. But it is useful. Heck sometimes it’s even been a senior dev to some of my queries, because I need the knowledge I don’t gave.

But it’s not a human. I can’t coach it and it’s not my friend. Boundaries and purpose

It's a new hammer

but one that is improving at an exponential pace and is developing capabilities to use itself with increasing reliability

It's easy to look at AI and draw a simple analogy to existing tools, because in most cases it is used as a tool, but the properties of intelligence and its ability to make things in the world is very unique and not comparable to any other tool.

All tools are useful because they require intelligence to use, and the tool magnifies the aim of intelligence. When the tools become intelligent themselves, certain recursive feedback loops will start to appear. Simply look at the quality of AI code outputs from 2 years ago compared to today.

> Boundaries and purpose

This can't be emphasized enough!