Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore practical strategies for implementing generative AI at the edge without relying on cloud infrastructure in this 15-minute deep-dive conversation with IDC Research Director Rob Tiffany. Learn how organizations are building private language models, optimizing them for performance, and deploying intelligence across diverse edge environments from factory floors to mobile devices. Discover a comprehensive framework for centralizing enterprise knowledge, compressing and quantizing models for on-device execution, and aligning model sizes with specific hardware tiers including gateways, kiosks, ARM devices, phones, and AI PCs. Examine hybrid architectures that maintain critical inference locally while leveraging private cores for heavy reasoning tasks. Gain insights from real-world examples, including an offline kiosk avatar that operated without cloud connectivity, along with field-tested approaches for balancing memory, latency, cost, and privacy considerations. Understand the economic drivers behind hybrid inference strategies, including cloud token costs, egress fees, and latency penalties that push organizations toward on-premises and on-device model deployment. Access practical guidance for developing generative edge AI roadmaps across various applications including standard operating procedures, field service assistance, part lookup systems, customer support, and internal policy Q&A systems.
