Uploaded May 2026 | Updated September 2026, 2 weeks ago

---

Enterprise AI agents need more than a model. They need open models, reliable frameworks, secure runtimes, and infrastructure that can support long-running, autonomous workflows.

In this video, LangChain shares how developers can build enterprise AI agents using NVIDIA Nemotron models, NVIDIA inference endpoints, NeMo Fabric, LangChain, LangGraph, Deep Agents, OpenShell, and DGX Spark.

What you’ll learn:
How NVIDIA Nemotron models and inference endpoints integrate with LangChain, LangGraph, and Deep Agents
How NeMo Fabric helps make agents faster and more reliable
Why secure runtimes like OpenShell matter for coding agents
How containerized runtimes and sandboxes support more autonomous agent workflows
Where DGX Spark fits into GPU-accelerated agent development

➡️ Learn more about NeMo Fabric: developer.nvidia.com/nemo-agent-toolkit
➡️ Run OpenShell: build.nvidia.com/openshell
➡️ Explore DGX Spark: build.nvidia.com/spark/nemotron