I Forgot How to Code After Using AI Agents
I Forgot How to Code After Using AI Agents
You sit down for a system design interview. The interviewer asks you to sketch out an API. Six months ago, this was automatic - endpoints, data models, error handling flowing from memory. Now your brain does something new: it waits. It expects to delegate. The code is not gone from your memory, but the retrieval pathway has changed.
The Cognitive Shift
Heavy use of AI coding agents changes how your brain approaches problems. Instead of thinking "how do I implement this," you start thinking "how do I describe this so the agent implements it." The skill shifts from writing to specifying. From doing to directing.
This is not inherently bad. Directors are valuable. But it creates a dependency that becomes visible the moment you cannot delegate - in an interview, during an outage, or when the AI service is down.
What Atrophies
The specific skills that fade with heavy AI usage:
- Syntax recall. You know the concept but cannot remember the exact function name or parameter order.
- Debugging intuition. You used to scan code and feel where the bug was. Now you paste the error and let the agent find it.
- Implementation planning. Breaking a feature into steps used to happen in your head. Now you describe the feature and let the agent figure out the steps.
What Gets Stronger
Not everything atrophies. AI-heavy developers often get better at:
- Architecture and system design (ironically, the interview topic)
- Code review and pattern recognition
- Describing requirements precisely
- Evaluating tradeoffs between approaches
Finding the Balance
The solution is not to stop using AI agents. The solution is deliberate practice without them. Write code by hand occasionally, the way a pilot practices manual landings. Keep the muscle memory alive even when you usually fly on autopilot.
Your AI agent is a tool. Make sure you can still function without it.
Fazm is an open source macOS AI agent. Open source on GitHub.