Microsoft Foundry & NVIDIA NeMo: Build Custom AI Agents

San Francisco, CA – Microsoft is significantly expanding its artificial intelligence capabilities, unveiling a series of updates to its Foundry platform and Azure AI infrastructure at NVIDIA GTC 2026. The announcements underscore the company’s commitment to providing a comprehensive, end-to-end platform for building, deploying, and operating AI agents and “Physical AI” systems, moving beyond simply training large models to enabling real-world applications. This push aims to position Azure as a central hub for the entire AI lifecycle, from initial development to large-scale production.

At the heart of Microsoft’s strategy is Foundry, which the company now describes as the “operating system for AI.” Built on Azure, Foundry integrates models, tools, data, and observability into a unified environment specifically designed for production-grade AI agents, rather than solely for experimental purposes. The company is expanding Foundry’s capabilities across Foundry Agent Service and NVIDIA Nemotron models, offering enterprises a more streamlined and production-ready stack for agentic AI. This move reflects a broader industry trend toward operationalizing AI, making it a reliable and scalable component of business processes.

Foundry Agent Service and NVIDIA Nemotron Integration

A key component of the update is the general availability of the next-generation Foundry Agent Service and its Observability features within Foundry Control Plane. Foundry Agent Service empowers teams to develop AI agents capable of reasoning, planning, and executing actions across various tools, data sources, and workflows. Crucially, Foundry Control Plane provides developers with end-to-end visibility into agent behavior in production environments, enhancing both developer productivity and enterprise trust. This level of observability is critical for ensuring AI systems operate predictably and responsibly.

Microsoft is also integrating NVIDIA Nemotron models into Foundry. This allows organizations to directly build and deploy specialized agents leveraging these open models. The combination of Foundry’s operational framework and Nemotron’s capabilities aims to accelerate the development of customized AI solutions tailored to specific business needs. According to Yina Arenas, Corporate Vice President of Microsoft Foundry, the expanded capabilities are designed to assist customers “build, deploy and operate production-ready AI agents on NVIDIA accelerators.” Microsoft Blog

Azure AI Infrastructure Optimized for Inference

Beyond Foundry, Microsoft is also bolstering its Azure AI infrastructure, specifically optimizing it for inference-heavy, reasoning-based workloads. A significant milestone is Azure becoming the first hyperscale cloud provider to power on next-generation NVIDIA Vera Rubin NVL72 systems. These systems are designed to accelerate AI inference, the process of using trained models to make predictions or decisions. Faster inference speeds are essential for real-time applications like fraud detection, personalized recommendations, and autonomous systems.

The Vera Rubin NVL72 systems represent a substantial investment in AI infrastructure, signaling Microsoft’s commitment to providing cutting-edge resources for its Azure AI customers. This infrastructure upgrade is expected to significantly improve the performance and efficiency of AI applications running on Azure, enabling more complex and demanding workloads. The move also positions Microsoft as a leader in the rapidly evolving AI hardware landscape.

Physical AI and Integration with NVIDIA Omniverse

Microsoft is also deepening its integration across Foundry, Microsoft Fabric, and NVIDIA Omniverse libraries and open frameworks to support “Physical AI” systems. Physical AI refers to AI systems that interact with the physical world, such as robots, autonomous vehicles, and smart factories. The integration with NVIDIA Omniverse, a platform for building and simulating virtual worlds, allows for the creation of digital twins – virtual representations of physical assets – that can be used to train and test AI models in a safe and controlled environment before deployment in the real world.

This integration is particularly valuable for industries like manufacturing and logistics, where optimizing physical processes is critical. By simulating real-world scenarios in Omniverse, companies can identify potential issues and improve the performance of their AI-powered systems before they are deployed in production. The ability to seamlessly transition between simulation and real-world operations is a key advantage of this approach.

Corvus Energy Case Study

Microsoft highlighted Corvus Energy as an early adopter of Foundry, demonstrating the platform’s real-world impact. Corvus Energy is utilizing Foundry to replace manual inspection workflows with agent-driven operational intelligence across its global fleet. Microsoft Blog This application showcases how AI agents can automate tasks, improve efficiency, and provide valuable insights that were previously unavailable. The success of Corvus Energy serves as a compelling example of the potential benefits of Microsoft’s AI platform.

The Broader Implications for AI Development

These announcements from Microsoft at NVIDIA GTC 2026 represent a significant step forward in the evolution of AI development. By providing a comprehensive platform that spans the entire AI lifecycle, Microsoft is aiming to lower the barriers to entry for organizations looking to adopt AI. The integration of NVIDIA technologies further enhances the platform’s capabilities, providing access to cutting-edge hardware and software. MSFT News Now

The focus on production-ready agents and Physical AI also reflects a growing recognition that AI’s true potential lies in its ability to solve real-world problems. As AI systems become more sophisticated and integrated into our daily lives, ensuring their reliability, scalability, and trustworthiness will be paramount. Microsoft’s investments in Foundry and Azure AI infrastructure are designed to address these challenges and pave the way for a future where AI is a powerful force for innovation and progress.

The expansion of Microsoft Foundry capabilities, coupled with the optimized Azure AI infrastructure and advancements in Physical AI, positions the company as a key player in the ongoing AI revolution. The emphasis on end-to-end solutions and real-world applications signals a shift from experimentation to operationalization, making AI more accessible and impactful for businesses across various industries. Technology Magazine

Looking ahead, Microsoft will likely continue to invest in AI infrastructure and platform development, focusing on areas such as edge computing, responsible AI, and AI-powered automation. The next major update regarding Azure AI and Foundry is anticipated at Microsoft Ignite in late 2026, where further details on new features and partnerships are expected to be revealed.

What are your thoughts on Microsoft’s latest AI advancements? Share your comments below and let us know how you see these technologies impacting your industry.