The Stack Engine: A headless WASM/WebGPU implementation. I’m currently using custom flags and selective streaming to bypass standard browser memory and execution limits.
Physics: I’ve ported Jolt (Guerrilla Games) but modified the solver to mimic the original Blam! engine (Bungie/Halo 1-3). It successfully replicates that specific "physics chaos" (unstable vehicle momentum, grenade jumping) that modern rigid-body sims often feel too "clean" to capture.
The "Vibe Console": A React-based no-code layer on top of a Godot web-editor clone. It uses an LLM-backed (Claude) interface to allow real-time engine modifications via natural language—essentially mapping prompts to engine dials and knobs.
The Vision I want to build a platform where casual/semi-competitive party games launch instantly upon page load. No installs, no long loading bars. I’m launching with three open-source "proofs of concept":
A Halo 2-inspired multiplayer arena.
A Simpsons Hit & Run style driving sandbox.
A 3D "Mousetrap" physics runner.
Why I’m posting here I’m a bootstrapped guy with a previous 10x SaaS exit and a light background in PDEs. While I’ve "vibe-coded" the current architecture into existence, I’m looking for a lead systems engineer who lives and breathes low-level optimization.
I have the capital and gtm know how, but I need a partner who can help me harden the WebGPU pipeline and ensure the "Vibe Console" doesn't create unmaintainable technical debt.
I’d love to hear the community’s thoughts on:
The viability of WebGPU for high-fidelity physics at scale in 2026.
The "Blam!" physics approach—is there a better way to simulate retro-instability than modifying Jolt's constraints?
If you’re a cracked engineer who wants to build the next-gen web gaming substrate, let’s talk.