James Sturtevant

@aspenwilder.bsky.social

right clicking and viewing source

Old rule of thumb: new tech must be 10x the status quo to catch on. Hyperlight: "How about 200x?" This revolutionizes anything needing: * hardware isolation (security), and * ultra-fast startup times (performance-sensitive / massively scalable, on-demand functions). Agentic AI workflows? What else?

squillace@squillace.bsky.social · 3w ago

Great little post on the @unikraft.com blog about using unikernels with the @cncf.io project #hyperlight: unikraft.org/blog/2026-06... like that boot-to-main()! (and that VM is < 1MB, so density.... which means cost down)

Absolutely! We are about to ship an Arazzo-based workflow engine with an option to scale out the runners with Hyperlight. It's also interesting for a secure-by-default multi-tenant architecture we've been slowly boiling for the last couple of years.

Running #Wasm components inside of a hypervisor and being able to tailor the exact interface between the hosted application and the trusted compute base enables platform builders unequaled ability to tune precisely the access provided to untrusted application code.

squillace@squillace.bsky.social · last yr.

Something I can't wait to talk about at #wasmio and @rejekts.io and at #kubecon: how to run #webassembly components at scale but with efficient hardware protection using #hyperlight. aka.ms/hyperlight-w...