"Everybody’s Slice of the Pie Became Smaller" - Why GTA Veteran Obbe Vermeij Left Rockstar

For Obbe Vermeij, the turning point wasn’t controversy. It wasn’t creative conflict. It wasn’t even burnout. It was scale.

In Part 1 of our exclusive interview, Vermeij revealed that Grand Theft Auto III began development before it was formally approved. In Part 2, he explained how subtle technical illusions - drifting litter, artificial headlights, a barely perceptible sun dim - made early GTA cities feel alive.

This final installment turns to something more personal: why he left Rockstar after GTA IV and why he chose to build something radically different.

Vermeij’s last Rockstar title was Grand Theft Auto IV. He left shortly after it shipped in 2009. There was no public dispute, no dramatic exit. The reason, he says, was structural rather than personal.

The team kept growing and everybody's slice of the pie became smaller. During GTA III I could work on something and see it in the game. By GTA IV there were so many people that it was harder to have that kind of impact.

In the early days, a developer could prototype an idea and feel its presence in the finished product. By the time GTA IV rolled around, Rockstar North had become a vastly larger operation. That scale allowed for unprecedented fidelity and ambition, but it also changed the daily rhythm of development.

I wasn't unhappy. It was more that the thing I enjoyed most about the job was becoming less and less possible. When you go from a team of 30 to a team of hundreds, the nature of the work changes. You spend more time in meetings and less time actually making things.

For Vermeij, the most rewarding part of development was direct impact. In the early GTA days, a single developer could prototype a feature and see it reflected almost immediately in the game world. As the teams grew, responsibilities narrowed and specialization increased. That shift is common across large-scale productions, but for someone who thrived on broad technical ownership, it marked a turning point.

At the same time, the tools of development were evolving rapidly.

When I started, you had to know assembler, you had to write your own physics, your own shaders. Those skills are basically obsolete now. Modern engines handle all of that. A young team today can make something that looks incredible without knowing any of that stuff.

Rather than viewing that change as a loss, Vermeij sees opportunity. Modern engines, middleware, and development tools have dramatically lowered the barrier to entry. What once required a full engineering department can now be accomplished by a small, focused team. That democratization is part of what made his current project possible.

That project is Plentiful.

Plentiful is a nature sandbox built around systems instead of spectacle. Players shape terrain, guide a small tribe, and observe how ecosystems respond. Water flows dynamically across the landscape. Terrain deforms. Plants grow, animals graze, and the world reacts to player intervention.

Plentiful is what happens when I stop making games for millions of people and start making something for myself. It's a nature simulation where you shape the land and watch a tribe figure out how to live on it.

Unlike a blockbuster open-world title built by hundreds across multiple studios, Plentiful is deliberately intimate. It draws inspiration from classic god games, but is built with modern tools that allow one developer to experiment quickly and iterate freely.

The game currently has a playable demo available on Steam and recently appeared during Steam Next Fest, where players could experience its evolving systems firsthand.

Vermeij’s comments on the future of GTA remain measured rather than critical. When asked about the next evolution of online architecture, he highlighted the scale of the challenge rather than speculating recklessly.

The online stuff is incredibly hard. The current GTA Online is already impressive but it's really more of a glorified network game in terms of architecture. Getting to the point where you have a true persistent world with that many players is a massive engineering challenge.

He was quick to acknowledge the teams responsible for that infrastructure.

Rockstar San Diego wrote the rendering part of the engine and Rockstar North wrote most of the network code.

On artificial intelligence tools, his view was similarly pragmatic. AI, he suggests, is a tool, not a replacement for design thinking.

AI is useful for coding. It can help with boilerplate stuff, debugging, things like that. For the kind of work I did back in the PS2 era it wouldn't have been relevant because everything was so low level and custom.

For Vermeij, the contrast between AAA scale and indie autonomy isn’t about right or wrong. It’s about preference. Some developers thrive in massive collaborative environments. Others prefer complete ownership of the systems they build.

I'm having more fun making games now than I have in years. I don't miss the big teams. I miss the people, but not the process.

Plentiful represents that preference in action. A project shaped entirely by one vision, where experimentation happens daily and iteration is immediate. It is the logical endpoint of a career that began in small teams building foundational GTA systems and evolved into global productions.

If you’re curious about what that philosophy looks like in practice, you can find Plentiful on Steam and try the demo for yourself. Also, if you want to follow Vermeij's ongoing commentary on game development and GTA history, he's active on X (formerly Twitter) at @ObbeVermeij, as one of the most candid former Rockstar developers willing to talk about how these games were actually made.

This concludes our three-part exclusive interview series with Obbe Vermeij.