Lightning Talk: AI at the Edge: ONNX Inference in WASM on Featherweight K0s
summary
Edge deployment runs ONNX inference inside WebAssembly using WASI on ONNX in WASM with a lightweight k0s single-binary Kubernetes cluster. Two Raspberry Pis per solar-powered buoy handle sensor collection and CPU-only inference, with the model output written as JSON that toggles GPIO-controlled pumps to dose nutrients. The system is low-cost, self-healing, and deployed on about 1,200 buoys worldwide, demonstrating practical, offline edge AI that actually works.
AI-written summary · about
related talks
10mLightning Talk: Scaling AI With Wasm and Edge Computing
36mApplied AI on the edge
05mLightning Talk: Running Energy-Aware AI Inference on Edge Kubernetes With Kepler
12mLightning Talk: Orchestrating Machine Learning on Edge Devices...
32mWebAssembly Based AI as a Service on the Edge with Kubernetes
33mWebAssembly based AI as a Service with Kubernetes
33mBringing AI to the Heterogeneous Edge with WebAssembly & ONNX
06mLightning Talk: Deploying AI Solutions on the Edge with Containers