Edge AI: Driving Agentic Workflows and Productivity
summary
Claims on-device AI with the AI Suite SDK running computer vision and small LLMs, cutting cloud tokens and latency while keeping data local. Demonstrations include product recognition supporting 200,000 SKUs on handhelds, a multi-barcode video barcode finder, delivery-photo compliance with real-time feedback, and voice transcription using Whisper tiny/small models. A set of pre-trained enablers and blueprint apps feed into Zebra Companion, an agentic assistant that currently keeps a human in the loop and pushes only complex tasks to the cloud.
AI-written summary · about
chapters (11)
- 0:00Edge AI introduction
- 1:25AI for frontline workers
- 2:50Real-time decision making
- 4:15AI developer tools
- 6:03Digitizing workflows
- 7:07Knowledge agents for onboarding
- 8:54On-device AI models
- 9:35Agent collaboration
- 11:29Task reallocation and trust
- 14:03AI assistant demo
- 16:16Future of agent UI UX
related talks
29mEdge AI: Driving Agentic Workflows and Productivity
48mEmpowering the frontline: Develop high-performance AI applications seamlessly for Zebra devices
20mImplementing Vision AI in Enterprise Android Apps
30mAI in the Cloud
10mRealizing the Power of AI at the Edge
19mAI, day 3
14mArm: One Platform for Agentic AI, from Cloud to the Edge
06mAI in production: The droids you’re looking for