Session Catalog
March 15-19, 2026 | San Jose, CA
Gain a holistic understanding of AI factory networking, exploring how key communication methods between GPUs, racks, and storage shape overall AI performance and scalability.
Learn techniques for updating and extending large language models with new knowledge, including fine-tuning approaches and knowledge injection methods.
Explore how to build sophisticated AI agents that can process and reason across multiple modalities including text, images, and audio.
Master prompt engineering techniques to build effective LLM applications, including chain-of-thought, few-shot learning, and structured prompting.
Learn to build multi-agent systems for asset lifecycle management with observability and scalability in mind.
Learn best practices for deploying and optimizing AI inference workloads at scale using NVIDIA inference platforms.
Get started with GPU acceleration using CUDA Python to speed up your computational workflows.
Simulate, train, validate, and deploy robotics workflows using the comprehensive NVIDIA Isaac platform.
Learn to construct 3D data pipelines essential for training and deploying physical AI systems.
学习如何使用 NVIDIA Cosmos VSS 平台构建智能交通系统违章检测解决方案。
探索使用 verl 框架进行大规模语言模型强化学习的最佳实践和方法。
了解如何使用 PhysicsNeMo 加速机器人柔性触觉传感器的仿真和开发。
了解 NVIDIA AI Enterprise 全栈如何赋能企业构建 AI 工厂，重塑智能制造标准。
探索面向机器人灵巧操作任务的高效强化学习框架设计和实现。
介绍 ROLL 框架：一个高效易用的大规模 LLM 强化学习框架，集成 NVIDIA Megatron。
回顾 NVIDIA 十年来如何赋能创业公司在 AI 时代实现快速发展和创新。
探讨高速公路数字化转型中，如何利用云边协同和 VLM 大模型解决行业长尾场景问题。
探讨工业级具身智能技术的最新进展，以及如何从概念阶段走向实际应用。
分享构建面向大规模 Agent 强化学习的高并发沙箱系统的实践经验和技术方案。
介绍如何基于 NVIDIA 全栈技术构建代理式 AI 和物理 AI 的基础架构和解决方案。