Effort Asymmetry
When everyone else's work is your play.
Developers are starting to notice something strange. Building with AI agents doesn’t feel like coding anymore. It feels like playing Starcraft. Like directing units instead of typing syntax. The Vercel CTO spent his vacation building software for 10 hours a day and compared the endorphin rush to slot machines. People on Reddit are talking about “3 or 4 five hour sessions a day.” Gamers who used to spend weekends on Apex Legends are now spending them on vibe coding.
Same thing happened to me.
What Changed
I used to come home from work and face a choice: work on a side project, or play games to decompress. The side project always required willpower. Gaming was the default. Gaming usually won.
Now building with AI is the game. I’m not competing with gaming for my leisure time anymore—building replaced it. Same evening hours, zero willpower required. The resistance just disappeared.
It’s the same reward loop. Rapid iteration. Pattern recognition. The satisfaction when systems click into place. My brain doesn’t know the difference between this and gaming. I don’t track hours anymore. I don’t need to force myself to start.
The Compounding Effect
Here’s where it gets interesting. AI agents already give you leverage—ship in hours what used to take weeks. That’s one multiplier.
But if building also feels like play, you stack another multiplier on top. More hours, because you want to keep going. Multiple projects, because you’re not depleted. Background processing, because you think about it in the shower.
These don’t add. They multiply.
The person working 8 hours with AI tools is 3x faster than before. The person playing 16 hours with AI tools is 6x faster AND doing twice the volume. That gap compounds every week.
Naval Ravikant put it this way: “I’m just playing, for sixteen hours a day. If others want to compete with me, they’re going to work, and they’re going to lose.”
That’s always been true. But agentic AI made the gap wider.
The Overlap I Didn’t Expect
Here’s something I’ve been noticing. The skills that made me good at games are the same skills I’m using to direct AI agents.
Rapid iteration toward a goal. Comfort with failure. Pattern recognition. Enjoyment of complex systems. Managing multiple things at once. Knowing when to push and when to step back.
Gaming taught me all of this. And it turns out that’s exactly what building with AI agents requires.
Early on, AI coding tools were still engineer territory. You needed to know what good code looked like to get good code out. But that’s shifting. The bottleneck is moving from “can you code” to “can you direct, iterate, and ship.”
Gamers who never learned to code are starting to build things. I don’t think most people have connected these dots yet—that the grind mentality and systems thinking from gaming transfers directly to this new way of building.
The Real Asymmetry
This isn’t about avoiding burnout or lasting longer. It’s about giving more.
The person playing puts in more hours without trying. They think about it constantly—shower, walks, dinner. They run multiple projects because they’re not depleted. They’re in flow state instead of grind state.
Paul Graham wrote about this years ago: “If something that seems like work to other people doesn’t seem like work to you, that’s something you’re well suited for.”
If building feels like play to you, you’re going to outwork everyone still grinding through willpower. Not because you’re more disciplined. Because you don’t need to be.