An engineer I mentor said something to me a few weeks ago that I’ve been thinking about ever since.
For many months, they took every chance they got to roll their eyes at AI. They’d dismiss it in conversation, joke about it, and say pretty directly that they didn’t buy the hype. This wasn’t someone checked out. This was someone who loved the craft of writing code, had spent over a decade getting excellent at it, and wasn’t ready to pretend this shift felt comfortable.
Over time, I started seeing a change. They began asking better questions about prompts, tools, and model choice. They started experimenting more and talking more openly about what this might mean for their career.
Then one day they told me: “I do have good product sense and good taste. Maybe I’ll be ok.”
I didn’t say anything for a second. That’s a hard sentence to arrive at when your identity has been tied to code for years. But when they said it, I could tell they were starting to find their footing again.
The Grief Is Real
Here’s what I think gets missed in most of the discourse around AI and engineers: the craft wasn’t just a skill. It was an identity.
Programmers don’t just write code. They belong to a culture with a shared language and standards for what good software looks like. In A programmer’s loss of a social identity, Dave Gauer, writing about what he calls “heirloom programmers,” put it well: “A social identity isn’t about typing on a keyboard. It’s about belonging to a group.” When the work that anchored that identity starts getting automated, the loss feels existential.
In a Reddit thread about this shift, one commenter wrote: “What’s disorienting is realizing that the thing many of us tied our identity to wasn’t just a skill, it was a position of scarcity. AI isn’t erasing creativity or problem-solving. It’s erasing exclusivity.” That felt closer to what I’ve been seeing on teams. The craft still matters, but the scarcity around execution is fading fast.
And this isn’t just a programmer problem. Operations, legal, finance, anyone whose professional identity was built on mastery of a complex, learnable craft is going through some version of this. In I think I’m going through the five stages of grief with AI, Diana Lace Davidova wrote that she’s landed “somewhere between bargaining and acceptance” after moving core ops workflows to AI. Engineers are just further along on the timeline.
Your Team Is Somewhere on This Map
Kübler-Ross’s five stages were about grief, and they describe identity loss too. And if you manage a team right now, you’re probably seeing different people in different stages at the same time.
The denial shows up as dismissal: “AI-generated code is garbage, I’d never use it.” The anger shows up as resentment toward colleagues who do use it, or toward leadership for encouraging it. Bargaining sounds like: “OK, I’ll use it for boilerplate, but the real architecture is still mine.” Depression is the quiet withdrawal, the engineer who used to be the most engaged person in the room and now just ships tickets without much commentary.
And acceptance, when it comes, looks like that engineer. Not defeated. Reoriented.
As a manager, your job at this stage isn’t to talk people out of their grief. Recognize where each person is. Don’t rush them. Someone in the anger stage doesn’t need a pep talk about AI productivity, they need to feel heard. Someone in depression might need a 1:1 that has nothing to do with tickets.
The Fracture Inside Teams
What makes this harder is that teams don’t move through this together. You end up with people at completely different stages trying to build the same thing.
Gauer’s “heirloom programmers” are the ones who feel the loss of craft most acutely. In The Programmer Identity Crisis, Simon Højberg described it sharply: “Creative puzzle-solving is left to the machines, and we become mere operators disassociated from our craft.” For engineers who found deep meaning in the problem-solving itself, that’s not a small thing to grieve.
On the other end, you’ve got engineers who have leaned in completely, shipping twice as fast, treating AI as a multiplier. They’re also sometimes tone-deaf to what their colleagues are going through.
The tension this creates inside a team is real and worth naming. You have people at different stages, with different identities, trying to build together. If you don’t acknowledge that, the friction starts shaping how the team works.
What Comes Out the Other Side
Back to that engineer.
“I have good product sense and good taste.”
The new moat for engineers isn’t code authorship. It’s judgment: knowing what to build, why it matters, and whether what’s being built is actually good. The engineers who thrive in this phase are the ones who can spot bad direction early, translate fuzzy product intent into something valuable, and review AI output for correctness and quality. Taste, judgment, and systems thinking aren’t soft add-ons anymore. They’re core.
What You Can Do as a Manager
Create space for the grief. This might be the most important one. If your team senses that the official party line is “embrace AI and move fast,” the people struggling will go underground. They’ll perform acceptance while privately falling apart. That’s bad for them and bad for your team’s actual output.
Watch for people stuck in bargaining or depression. The transition has a real cost, and it’s not evenly distributed. Some people move through quickly. Others get stuck. An engineer stuck in depression can look like low engagement, missed deadlines, or quiet disengagement from code reviews. It’s worth a direct conversation.
Name the new value explicitly. If you only celebrate shipping velocity and lines of code, that’s what your team will optimize for. Start celebrating taste. When someone catches a bad design decision early, name it. When someone pushes back on a feature direction because they have a better read on user behavior, name that too. The things that matter now need to be visible.
Your job isn’t to convince anyone that AI is good. It’s to help your team find their footing on the other side of a real disruption.
That engineer who told me they have good product sense and good taste is going to be more than ok. They found the skill AI struggles to replace: deciding what’s worth building in the first place. The identity crisis is real, and so is the way through it. It takes time, and it helps to have a manager who can tell the difference between someone who’s stuck and someone who’s still on their way.