The Slow Bit Nobody Talks About

All my regression suites run for hours. Even headless. Even with parallelisation. Playwright spins up a full Chromium process with a GPU compositor, extension subsystem, accessibility tree, print pipeline, the lot. For headless automation where nobody is looking at the screen, 90% of that machinery is dead weight burning clock cycles.

Lightpanda is building a browser engine from scratch in Zig, specifically for headless automation and AI agents. Not a Chromium fork. Not a wrapper. A ground-up engine that strips out everything an agent doesn't need.

No GPU compositor. No extension runtime. Minimal rendering pipeline. Just DOM, JavaScript execution, network stack, and the bits you actually test against. The claim is much faster execution for the headless use cases that dominate agent workflows.

Why Zig

Zig gives them manual memory control without the footgun of C and without Rust's borrow checker slowing down iteration speed. For a browser engine where you need to manage thousands of DOM nodes, layout objects, and JS contexts simultaneously, that matters. The binary is small, startup is fast, and there's no garbage collector pause eating into your test execution time.

Where It's At

Early. The engine doesn't cover the full web platform yet. Complex SPAs with heavy framework magic might hit gaps. But the project is active (13,000+ stars, funded, hiring) and the architecture is sound.

I haven't switched my test suites to it yet. But I'm watching it closer than anything else in the browser tooling space. The current approach of "take a browser designed for humans staring at screens and run it without the screen" was always going to hit a ceiling. Lightpanda is betting that a purpose-built engine can blow past that ceiling. I reckon they're right, it's just a question of when the platform coverage gets there.

For a deeper comparison with every other browser tool in the ecosystem, see the Browser Tools series.