Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the complete journey from cutting-edge foundation models to production-ready, globally available APIs in this 18-minute conference talk from AWS re:Invent 2025. Discover the comprehensive roadmap and architecture required for model performance optimization, multi-region GPU-based infrastructure deployment, and the intricate mechanics of request prioritization, token accounting, and rate limiting. Learn how to build robust production services using an inference stack that incorporates advanced technologies including NVIDIA Dynamo, TensorRT-LLM, and Amazon EKS, while tracing the complete path from raw model weights to scalable, high-performance AI services. Gain insights from AWS Partner Baseten on implementing enterprise-grade AI inference solutions that can handle frontier AI models at scale.
