Option 1 (Recommended):
"I have a developer who was literally training his own AI replacement."
Option 2:
"Our client was training an AI to replace one of our developers — and his job was to help them do it."
Option 3:
"There's a point on this graph where the AI gets good enough to take your job. One of my developers hit that point six months ago. He's still here. Here's why."
I have a developer who was literally training his own AI replacement.
Our client was building an AI system, and Jonathan's job was to show it exactly what he was doing — walk it through his process, help it learn, basically teach the machine how to do his job better.
And here's the thing: if you plot this on a graph, there's an obvious intersection point. The AI capability curve goes up, the human necessity curve goes down, and somewhere in the middle they cross. That's the moment the company doesn't need you anymore.
Jonathan hit that point six months ago.
He's still here. Actually, he's more valuable now than he's ever been.
So what happened?
He did his job. He helped the AI learn. He was good at it. But then he got his hands on an LLM coding tool — and he started building everything in sight.
Someone mentioned in stand-up, "Hey, it'd be cool to visualize all our image processing issues in a graph."
Same day. Done. "How's this?"
Another developer said, "It'd be great if the system could auto-prioritize images."
By end of day, there was an MVP with a working UI sitting in Slack.
Jonathan didn't run from AI. He ran straight into it. And now he's the most productive developer on the team — by a mile.
Look, AI isn't going away. You can stick your head in the sand and hope it doesn't come for your job, or you can do what Jonathan did.
Don't compete with AI. Multiply yourself with it.
The developers who figure this out aren't getting replaced — they're becoming irreplaceable.