Fast and flexible inference on open-source AI models at scale | BRK117

Run open-source AI models of your choice with flexibility—from local environments to cloud deployments using Azure Container Apps and serverless GPUs for fast, cost-efficient inferencing. You will also learn how AKS powers scalable, high-performance LLM operations with fine-tuned control, giving you confidence to deploy your models your way. You’ll leave with a clear path to run custom and OSS models with agility and cost clarity. To learn more, please check out these resources: * https://aka.ms/ignite25-plans-AIAppsCosmosDB 𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀: * Mehrdad Abdolghafari * Cary Chai * Sachi Desai 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻: This is one of many sessions from the Microsoft Ignite 2025 event. View even more sessions on-demand and learn about Microsoft Ignite at https://ignite.microsoft.com BRK117 | English (US) | Innovate with Azure AI apps and agents, Azure Container Apps, Azure Kubernetes Service (AKS) Breakout | Intermediate (200) #MSIgnite, #InnovatewithAzureAIappsandagents Chapters: 0:00 - Use cases: hybrid model architecture, LLMS agents, data boundary control 00:09:09 - Introduction to GPU-intensive workloads like physics and video processing 00:11:47 - Docker Compose for AI agents and simplified cloud deployment 00:16:00 - Live testing of the dashboard generator and log streaming visualization 00:20:31 - AKS investment areas: scale, security, cost optimization and AI support 00:25:04 - Enhanced workload scheduling and configuration for AI workloads 00:30:25 - Inference traffic management using Gateway API and Ignite demo preview 00:35:11 - RBC’s CI/CD pipeline accelerating secure GPU resource provisioning 00:38:01 - RBC strategy: building Canada’s largest AI farm within compliance boundaries