The God Mode Feeling
For execs and knowledge workers who just vibe coded an app and the sensation that followed
I was building a personal knowledge system, a local adaptation of Nate B. Jones’ Open Brain (OB1) framework, and somewhere around hour three, it worked. Not “mostly worked.” Not “I think this is right.” It worked. Claude had predicted the build would take about three hours, and it did, and the thing was running, and I had built it.
The feeling that followed was not what I expected.
It hit like a full head and body high, briefly disorienting, overwhelming in the best possible way, lasting well into the next hour with that voice running in the background: oh my god, I can’t believe it actually worked. I had expected satisfaction. I got something closer to euphoria.
I didn’t have language for it at the time. I do now.
What just happened to your brain
For most executives, building something has always been an act of direction. You define it, resource it, review it, approve it. The making happens elsewhere. What AI-assisted development does, for the first time for many of us, is collapse that distance entirely. You are the builder. And the brain, encountering a genuinely novel reward, responds accordingly.
Researchers at the Karolinska Institute found that dopamine release in the brain’s striatum increases measurably during deep creative engagement, not as a side effect but as the mechanism driving the experience itself. The brain reinforces the behavior in real time. When outcomes exceed prior expectations, what neuroscientists call a positive reward prediction error, the dopamine response is outsize. Your brain is updating its model of what you’re capable of. That update feels like electricity.
The variable reinforcement loop makes it stickier. A 2024 study in Nature Human Behaviour found that unpredictable rewards trigger roughly 3.5 times more dopamine release than predictable ones. Each prompt either works or it doesn’t, unpredictably, quickly. The near-miss, the prompt that almost works, is neurologically more compelling than one that always does. Researchers studying gambling behavior have documented this extensively: the brain treats a near-miss as almost winning, which sustains the loop more powerfully than consistent success. The prompt that fails at 11pm becomes the prompt you have to fix by midnight.
This is not a character flaw. It is a slot machine in your editor, and it was designed by evolution.
You are not unusual for feeling this. Y Combinator CEO Garry Tan admitted publicly to staying up until 5am because he was “so addicted” to his AI coding assistant.
The crash is proportional
Here is what nobody tells you: the higher the ceiling, the lower the floor feels on the way back.
When the session ends, when the thing is built and running and there is nothing left to fix, normal work returns. Email. Meetings. Decks. And for a day or two, those feel oddly gray. Psychology Today has documented this pattern in creative professionals as post-achievement depression: a sense of purposelessness after completing a long-standing goal, driven in part by the same dopamine system that fueled the pursuit. Researchers at NeuroLaunch describe it plainly: once you hit your target, the reward may be short-lived, leaving a temporary vacuum where the drive used to be. A BCG/HBR study from 2026 put clinical language to a related phenomenon, documenting mental fog, reduced concentration, and increased error rates following intensive AI work sessions, and coined the term “brain fry” for the state that follows.
The crash is proportional to the high. It is biological, it is temporary, and it is a signal that something went right.
What it is not: a reason to immediately open a new project and chase the feeling again. Research documented in the Gerlich (2025) cognitive offloading study of 666 participants found a strong negative correlation between AI usage intensity and critical thinking scores, with the sharpest drops occurring in the period following peak engagement. The transition from healthy flow to compulsive re-entry often starts here, at the end of the first high, when returning to the screen feels like the obvious move.
Touch grass
The part that gets lost in conversations about AI productivity is what the time savings actually buys.
If a build that would have taken a team three weeks took you three hours, the question is not what to build next. The question is what you have been postponing. The walk you have not taken. The dinner you rescheduled. The unstructured thinking that requires no screen.
The neurochemical argument is straightforward. Physical re-engagement, movement, daylight, unhurried time, is what restores baseline dopamine function after a peak. This is how the system works. The feeling will come back. The rebalance is what makes it sustainable.
Sources
Nate B. Jones, Open Brain (OB1) - https://github.com/NateBJones-Projects/OB1/tree/main
De Manzano et al. (2013), dopamine D2 receptors and flow proneness, Karolinska Institute — https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0038806
Nature Human Behaviour (2024), variable reward and dopamine release
BCG/HBR, “brain fry” study (Bedard et al., 2026) — https://hbr.org/2026/03/when-using-ai-leads-to-brain-fry
Psychology Today, post-achievement depression — https://www.psychologytoday.com
NeuroLaunch, happiness hangover —https://neurolaunch.com
Gerlich, M. (2025), AI Tools in Society: Impacts on Cognitive Offloading — https://doi.org/10.3390/soc15010006