~/talk

Building Serverless AI Apps with Spin and WebAssembly

on YouTube 25m
Talkat Cloud Native AI Day Europe 2024 · on Tuesday, 2024-03-19· at 16:25· Pavilion 7 | Level 7.3 | Room S03· [webassembly, serverless]

speakers

name:Matt Butcher
website:http://technosophos.com
name:Radu Matei
website:https://radu-matei.com/

resources

video:https://youtu.be/OjclCGazSdg
slides:none on file

summary

Spin compiles small serverless handlers to WebAssembly so a 2.4 MB wasm module running a LLaMa2 Chat LLM starts in ~0.51 ms locally, swaps transparently from M1/metal to a remote A100/CUDA without rebuilding, and runs on Kubernetes via spin cube while the host assigns fractional GPU slices for just-in-time inference. Demo shows a TypeScript HTTP handler calling a hosted infer() API, achieves fast cold-starts and higher GPU utilization versus multi‑GB container images.

AI-written summary · about

chapters (5)

  1. 0:00Intro
  2. 1:40What is WebAssembly
  3. 8:43What is Spin
  4. 15:43Demo
  5. 22:56Questions

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.