AI workloads are moving closer to where data is generated — but how close can we really get? This webinar explores the evolving landscape of deploying advanced AI models, including computer vision (CV) and large language models (LLMs), on embedded and edge devices. We’ll cut through the hype and address the practical challenges, current solutions, and architectural trade-offs involved in getting these models to work reliably and efficiently at the edge.

Topics covered will include:

• The current state of LLM and CV model deployment at the edge
• Practical limits of model size, compute, and memory on embedded devices
• Deployment trade-offs: cloud, edge, or hybrid approaches
• Toolchains, model compression, and quantization strategies
• Real-world pitfalls: thermal limits, power budgets, latency, and data privacy
• What's realistic today — and what might be possible tomorrow

Whether you're an embedded engineer exploring AI, or an ML practitioner eyeing the edge, this webinar will help you understand the boundaries and opportunities in this fast-changing space.