ldraw-nova: AI agents build LEGO models by writing Python geometry
In ldraw-nova, models write a plan and a Python generator instead of raw coordinates, but nothing checks physics or stability.
Netlify Edge Functions on Firecracker microVMs: 5x faster warm calls
Netlify's heavier Firecracker sandbox won because it removed a network round trip; count hops and track cold start rate before blaming the runtime.
Huseyin Kaplan · 1 min read
Netlify moved Edge Functions from V8 isolates to full Firecracker microVMs. Warm invocations got about 5x faster.
The heavier sandbox won because the old path left their network. Every matching request went out over the internet to a hosted execution service and came back. Now it runs on a compute node inside their own edge, around 5 ms median instead of a few dozen.
The write-up is honest about the costs:
- Sticky routing keeps a VM warm, and it's also how hot spots form. Past a threshold, they spread one service across several nodes.
- Each VM serves a fixed number of requests, then shuts down.
- About 1 in 100 invocations is cold, at roughly 9 ms.
Before you blame the runtime for edge latency:
- Count the network hops before your code runs.
- Track cold start rate, not only cold start time.
- Recycle workers on purpose.
The sandbox got heavier. The path got shorter.
