Bootstrapping edge infrastructure for AIoT applications with open source software

on YouTube ↗19m

" summary

Claim: developers can bootstrap edge AIoT infrastructure on COTS hardware using open-source components and a few commands. The stack uses MicroStack in four commands to stand up a compact OpenStack, MicroK8s plus Kubeflow in two commands to deploy end-to-end ML pipelines (GPU enabled), MAAS for provisioning, and Ubuntu Core with snapcraft for OTA updates and automatic rollbacks. The setup runs on laptops and even Raspberry Pi class devices, enabling local training, inference and fleet updates.

AI-written summary · about

" chapters (4)

1. 0:00Start
2. 2:53Empowering innovations witn open-source
3. 11:45Open source software stack for Edge Al
4. 16:51Ubuntu Core and Snapcraft

" related talks

• 33mThe Case for Open Source in Autonomous AI Systems in the EdgeTina Tsou, Sujata Tibrewala · AI.dev: Open Source GenAI & ML Summit North America 2023
• 17mMicroStack: Putting OpenStack at the edgeTytus Kurek · OpenInfra Summit 2020
• 10mWhat Powers Frontier IntelligenceThierry Carrez · KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026
• 11mKeynote: Open Source Stack Empowers the AI EraDr. Xiong Wei · Open Source Summit Europe 2025
• 33mEdge Computing using OpenStack Cloud for IOTRohit Londhe, Nilesh Chandekar · DevConf.CZ 2020
• 34mOpen and Neutral Edge Computing Architecture on Heterogeneous DevicesGavin Lu · Open Source Summit + Embedded Linux Conference North America 2019
• 11mOpen Source for the AI EraJonathan Bryce, Horace Li · KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026
• 49mEnable IoT with Edge Computing and Machine LearningJared Rhodes · Update Conference Prague 2018