We’ve all seen the posts:
“Tried AI on a real task—deployed bad code, cost us $800.”
“The model couldn’t complete the ticket. It’s not ready.”
But when I read these, my reaction isn’t, “AI’s not there yet.” It’s more like:
You’re using a nail gun like a hammer.
The Tool Isn’t Broken; You’re Using It Wrong
Engineers are trying to use AI like it’s a junior dev:
Hand it a Jira ticket, wait for a PR.
When it doesn’t work, the takeaway becomes: “AI coding tools are overhyped.”
But that mindset misses the point. AI isn’t a person. It doesn’t reason like you. It doesn’t infer unspoken context. And it doesn’t deserve to be judged by how well it mimics your workflow.
If you treat it like a drop-in replacement for your team, of course it’ll fall short.
But if you learn what it’s actually good at, and evolve your workflows around that, you start to see where the leverage is.
Workflow Innovation
The best engineers I know don’t dismiss a new tool when it doesn’t slot neatly into existing processes.
They ask: “What can I do now that I couldn’t before?”
Some ideas:
- Use AI to write spike docs or plans before writing code.
- Break complex tickets into small, repeatable prompts the AI can reliably tackle.
- Ask it to reason about incidents or compare historical failures while you investigate.
- Pipe your repo into a local context window and use it as a thought partner—not a code generator.
In short: the best engineers aren’t using AI to do their work.
They’re using it to change the nature of their work.
The Gap Will Widen
AI won’t replace engineers. But it will amplify the best ones.
The 10x engineer of tomorrow isn’t just faster—they’re multiplied by AI.
This will widen the gap:
- Great engineers will deliver 10x the impact—then 100x—as they master these tools.
- Average engineers will stall, stuck asking why AI can’t “just do the ticket.”
And when VC-backed products turn today’s AI experiments into tomorrow’s point-and-click tools, the playing field will flatten. Everyone will have access to the same capabilities.
Your only edge will be how well you understand the tools… before everyone else does.
It’s Not About the Nails
If you still think your job is to hammer nails faster than the next engineer, you’re already behind.
The goal isn’t to hammer faster. The goal is to build better houses.
And these new tools?
They’re not threats. They’re upgrades.
If we let go of old workflows and lean into new leverage, we don’t just stay relevant… we lead.
Discover more from zach wills
Subscribe to get the latest posts sent to your email.