Game Engine

If I wanted to vibe-code a game, which game engine should I choose? Which game engine is the best for vibe coding?

Unity

The first engine that came to mind was Unity. I know C#, it is popular, there are a lot of resources to learn from, and in general it is indie-friendly.

On the other hand, Unity has a track record of bad behavior toward the community and controversial licensing decisions. The most obvious example is the Runtime Fee controversy: Unity introduced a per-install fee in 2023, then cancelled it in 2024 after strong backlash from the developer community.¹

But the main problem for me was different: Unity is editor-first. LLMs cannot work with an editor effectively, even if some editor tools are exposed through MCP. I spent a couple of days on it before giving up — the editor-first workflow really did block the agent in practice, MCP bridges and all.

For the same reason I gave up on Unreal Engine — though there it was even simpler: just opening a hello-world project took about ten minutes, which is incompatible with any kind of fast iteration loop, agent-driven or otherwise.

Godot

Although Godot also has an editor, I found it easier to create the world with code. It is not ideal, but it works. Godot also has a strong, mature ecosystem and is open source.

Adopted on 2025-08-28; still the primary engine eight months later.

Bevy

This is the engine I fell in love with. It does not have an editor, and games made with it can look fantastic.

There are problems, though:

• it is written in Rust, and I do not know Rust well enough yet
• it is still very young, and the API is not stable
• breaking changes make agents struggle with different API signatures
• it is hard for the agent to consistently write compilable code
• compilation is very slow

Tried it for real over five days (2025-10-21 → 25): wrote ~69 Rust files across five crates, wired a ZeroMQ bridge to the C# core, even deleted the Godot project at one point. Then put it back.

So, for now, I decided not to work with it. But I will definitely check it out again later.

CryEngine, Stride, and others

There are also less popular game engines. They might be great, but most of the time LLMs do not know them very well and struggle to write useful code for them.

I did briefly try Stride (2025-10-20 → 21) — got as far as importing fonts before agents stalled on the asset pipeline.

My own engine

One of the solutions proposed by ChatGPT was to not use an engine at all, but to use something like Avalonia + Silk.NET. This could actually work, but it would require more experience from me.

I actually tried this once: a custom C# engine called Forge, vibe-coded in a single day (2025-10-30), going from milestone 0.1 (bootstrap) through 1.0 (production-ready) across 83 files — rendering, UI, animation, shadows, streaming, packaging. Worked well enough as a tech demo. Did not survive contact with the actual game.

So after testing all these options, I decided to go with Godot.

I am also trying to write the core of the game in an engine-agnostic way, with ports to concrete engines. In theory, this means I could switch engines later. In practice, I do not believe it would be easy.