Part 2: Beginners, Build Experience with AI — The Ones Who Grow Fastest Are Beginners

Series: “Careers & Skills in the AI Era” (5 parts)

1. How Not to Let AI Take Your Career
2. Beginners, Build Experience with AI (this article)
3. Make AI Your Student and Escape the Intermediate Trap
4. Grow Leaders and Managers with an AI Team
5. Those Who Grow AI Win

Introduction

Last time, we reframed the AI-era career question from “how to avoid being replaced” to “how to grow by partnering with AI.”

This time is the first stage: beginners.

“If AI answers everything, how can a beginner ever grow?”—I hear this a lot. But I think the opposite is true. This has become an era where beginners, precisely, can build experience the fastest using AI.

The Myth That “You Can’t Build Experience with AI Around”

First, let’s untangle a common myth.

If all you do is get answers from AI and copy-paste them, then sure, you won’t build experience—because you move neither your hands nor your head. But that’s a problem of how you use it, not a problem with AI.

AI dramatically lowers the “first hurdle” that used to stall beginners.

• Getting stuck on environment setup and unable to move forward
• Burning hours because an error message makes no sense
• Not even knowing where to begin

AI keeps removing these causes of “dropping out before building any experience.” In other words, the distance to building experience has shrunk.

Experiencing—and Failing—Together with AI

So how does a beginner build experience? The answer is simple: work alongside AI and fail alongside it. That repetition—iteration—is experience.

For example, these make excellent material:

• Building a dev environment for a new language from scratch
• Actually coding something small in that new language
• Trying a new version’s build procedure
• Touching infrastructure—setting up a container, installing an OS on a VM

Every one of these is the kind of thing where, on your own, you stall at the “first hurdle.” With AI, when you get stuck you can immediately try the next move. And what matters is not that it goes smoothly, but getting stuck together and getting unstuck together. The environment broke, the build won’t pass, the container won’t start—chasing down why becomes experience, exactly as it is.

The mindset here is to use AI not as a “source of answers” but as an accelerator for experiments. What used to take a day per attempt becomes ten attempts in an hour with AI. The number of iterations is the amount of experience (Simon Willison, who publishes daily experiments with AI, is a good model here).

Let AI Fail a Lot—and Watch

There’s a further move: deliberately have AI try a lot, and fail a lot.

On your own hands alone, the number of attempts is limited. But when you let AI and agents rack up the count, where the pitfalls are and what tends to break rises out of the pile of failures.

In fact, I get more information this way than working entirely on my own.

• Watch how the AI (agent) decides and where it stumbles
• Read the emitted logs and errors together with AI
• A feel for “this step is risky” or “this is environment-dependent” accumulates as experience

AI isn’t afraid to fail. So using AI as a “scout that fails ahead of you” makes your own experience grow fast.

From “Getting Answers” to “Working Alongside”

As a beginner, it’s fine to start by asking AI and getting answers. That’s an efficient entry point.

What matters is not stopping there. Once you get an answer, always run it yourself, change it a little, and try breaking it. Gradually shift your weight from “getting answers from AI” to “working alongside AI.” That shift is what turns a beginner into a fast grower.

Those who dump everything on AI never accumulate experience; those who make AI an experimentation partner watch their experience snowball. Using the same AI, the gap after a few months becomes surprisingly large.

Summary

• “You can’t build experience with AI around” is a copy-paste-based myth; the reverse is true depending on how you use it.
• AI lowers a beginner’s “first hurdle” and shortens the distance to building experience.
• Experience accumulates through iteration—working with AI and failing with it.
• The stickier the material—dev environments, coding, builds, containers/VMs/infra—the better the practice.
• Letting AI fail a lot and watching yields more information (pitfalls, gut feel) than going solo.
• Shifting from “getting answers” to “working alongside” is the fork in the road toward becoming a fast grower.

Next time: how to get past the intermediate wall that so many people hit once they’ve built experience. The keyword is “teaching AI.”

👉 Next: Make AI Your Student and Escape the Intermediate Trap