Engineer Stories

I used to feel proud after a hard debugging session. Even if it took three hours, even if I had to go for a walk and come back — there was something in that struggle that felt like me doing the work. Then AI came and I could ship features in an afternoon that used to take days. The output looked the same. The feeling didn't. I started noticing I couldn't debug things on my own anymore. Not slowly — completely. Like there was a blank where the instinct used to be. I took the quiz and scored 9. I cried a little. I'd been telling myself I was just tired. Started doing one hour a week with no AI. Just me and the problem. First week felt unbearable. Fourth week I could feel the muscle coming back.

The worst part wasn't the fatigue. It was that I stopped being curious. I'd used to read tech blogs at lunch, think about architectural decisions on walks, pick up a new language on weekends for fun. Then AI could answer my questions instantly, and I stopped asking. My curiosity didn't fade — it got replaced. Every question I would have looked up, I asked the AI. Instant answer, instant closure. No gap to sit in, no moment of not-knowing that made the answer worth finding. The quiz helped me see this pattern. I scored 8. Now I keep one 'stupid question' per week — something I won't ask AI, something I have to figure out myself. Feels embarrassing. Also feels like me again.

I got my first dev job 18 months ago. I learned to code with AI tools from the start — when something broke, AI explained it. When I didn't understand something, AI clarified it. Then my tech lead asked me to whiteboard the system we'd been building. I opened my mouth and nothing came out. I could navigate the codebase. I could explain individual pieces. I could not explain the whole. I scored 11 on the quiz. The 'junior engineers' page hit me like a truck. The skills I was supposed to be building were being built in the AI, not in me. I didn't have a mental model of the system — I had a route map through it. I'm now doing 'debug Sunday' — three hours with no AI, just me and a broken feature. It's humbling. It's also the only time I feel like I'm actually learning anything.

I scored 2 on the quiz. I'm doing fine — for now. But I recognized the signs from my teammates who aren't. I've been protecting one hour every morning before standup — no AI, just me and whatever I'm building. Started as an experiment. It's now the part of my day I protect most fiercely. The clearing-ai quiz helped me name what I'd been feeling in a small way: that the efficiency gains were coming from somewhere, and that somewhere was the slow erosion of something I didn't want to lose. I'm not burned out. I'm aware of what could cause it.

I'm watching my team burn out and I didn't have a name for it until I found The Clearing. We shipped a major feature last quarter. The velocity was incredible — AI tools handled so much of it. But when something broke in prod last week, my two senior engineers stared at each other for ten minutes before either could form a theory about what was wrong. Both reached for AI simultaneously. I scored 10 on the quiz. My team is Tier 3 collectively. The problem isn't that they're not trying — it's that the system is set up to erode exactly the skills they'll need when things go wrong. I'm building an 'explanation requirement' into our code review process. If AI wrote it, the author explains it without AI. Harder than it sounds. Worth it.

I took the quiz in January after my therapist asked me to. Scored 13. I'd been working 60-hour weeks for months. Shipping fast, feeling empty. My marriage was strained. I couldn't remember the last time I'd felt proud of something I built — not just shipped, but built, the way I used to understand it. The resource that helped most was the daily-ai-boundaries page. I started with 20 minutes of no-AI time before work. Gradually extended. It's May now and I'm down to 40 hours, I've taken two weekends completely off, and I remembered what it feels like to be curious about code again. I almost left the industry. I'm glad I didn't. The quiz didn't fix me — it just gave me permission to name the problem and start fixing it myself.

Remote work was supposed to be freedom. Then AI made 'always-on' the default. I started at 8am, AI helping me through every problem. Finished at 6pm, but the AI was still running in my head at dinner. Cognitive separation between work and life — something I fought hard to build as a remote engineer — started dissolving. Scored 9 on the quiz. The 'remote work AI fatigue' page described the thing I couldn't name: ambient availability. The AI is always there, so the work is always there. No natural endpoint. I now have a hard shutdown ritual — actual time, actual switch. Takes practice. The default mode network research from The Clearing helped me understand why the separation matters neurologically, not just emotionally.

I thought I had imposter syndrome. Turns out it was something different. I'd been telling myself: everyone else can keep up, I just don't belong here. But then I found The Clearing and took the quiz and the imposter-syndrome page said 'this is not imposter syndrome — this is functional change.' That distinction mattered more than I expected. Imposter syndrome says: you're wrong about yourself, fix your perception. AI fatigue says: something changed in how you work, here are the mechanisms. One is a defect. The other is a condition with a treatment. Scored 11. Started no-AI Fridays. Not because I want to prove something — because I want to know what I actually know.

The velocity is addictive. That's the part nobody talks about. I can ship in two hours what used to take me a full day. My manager loves it. My team loves it. I'm delivering more than I ever have. And somewhere in the last six months, the sense of authorship started draining out. Scored 8 on the quiz. The feeling: code appears, I ship it, I don't fully know why it works the way it does. Like signing a letter I didn't write. The 'ghost authorship' section of the site was the thing that finally named the experience I'd been trying to describe to my therapist for months. She gets it now — I sent her the link. We talk about the Explanation Requirement as a concrete practice.

My team of 8 engineers — all solid, experienced people — and none of them can debug a production issue without AI in front of them. I've started calling it 'debugging with training wheels.' Scored 10 on the quiz. I took it because I felt like something was wrong and I couldn't name it. The answer: collective skill atrophy at a team level, with AI as the accelerant. We're implementing 'no-AI Wednesday mornings' — two hours where the whole team works without AI assistance. Sounds absurd. Is actually working. The conversations that happen in those two hours — about why something behaves a certain way, about what we've seen before — are the things that used to make us a team.

I'm the kind of engineer who reads the changelog before updating dependencies. That same instinct is what brought me to The Clearing early. Scored 2 on the quiz. But the resource that helped wasn't the quiz — it was the research page. Understanding attention residue, the cognitive cost of context switching, the way AI breaks deep work. I now protect my focus blocks more intentionally because I understand the mechanism, not just the feeling. I think of myself as practicing prevention rather than recovery. The clearing-ai content gave me the vocabulary and the data to make a case to my manager for better async norms. Didn't have that before.

I'm one of those annoying people who reads documentation end-to-end before touching a new tool. Yes, it takes longer. No, I don't regret it. When AI coding assistants became mandatory at my company, I noticed the gap immediately: people who reached for AI first had no mental model of what the code was doing. I kept reading the docs. I kept building the map. Scored 3 on the quiz. My colleagues score higher — some much higher. The thing I've noticed: the people who are in trouble are the ones who can't calibrate their own uncertainty anymore. They can't tell when something is actually hard or when they just haven't thought it through. AI removes the calibration signal. My practice now: every time I use AI to solve something, I then try to solve it without AI. Not to prove a point. To keep the calibration signal alive.