Agentic AI Conference Sessions

AI agents are the new digital workforce, working for and with us. They can reason about a mission, create a plan, and retrieve data or use tools to generate a quality response. Data is the fuel for AI agents, but the magnitude and scale of enterprise data often make it too expensive and time-consuming to leverage effectively. For enterprises to thrive in the AI era, they must find a way to make use of all of their data. Join this session to learn about tools and frameworks to more easily build agentic AI systems that connectAI agents to a library of reusable tools that unlock your data, and drive efficiency gains across the organization.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

As a leading developer platform for generative AI orchestration, LangChain has been at the forefront of the journey of AI agents from experimental concepts to essential components in production systems. We'll cover key lessons learned from LangChain’s experience enabling customers to develop, deploy, and manage enterprise AI agents in production at scale, and will explore emerging technologies that will shape the future.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Enterprises struggle to implement generative AI, but are looking to transition to full-scale production for cost savings and new revenue models. Reliability, security, and scalability are crucial for mission-critical AI applications. We'll explore best practices for implementing AI strategies, highlighting AI's value and potential impact. Learn how customers leverage NVIDIA AI Enterprise, which includes NVIDIA NIM and NVIDIA NIM Agent Blueprints, to gain insights on challenges, lessons, and best practices for all stages of the AI journey.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Agentic AI is inspiring a paradigm shift from software-as-a-service to service-as-software. Agents combine reasoning, dynamic data retrieval and can access tools to drive outcomes. Teams of AI agents can work together, achieving results that are often far better than those of a single AI agent. With NVIDIA AI Blueprints, building blocks for agentic AI, we will show how enterprises can easily compose teams of agents, and securely deploy them with confidence.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Join us for a deep dive into how generative AI is being used to train foundation models at scale.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

Join NVIDIA technical product architects for an in-depth tutorial demonstrating how to build agentic AI pipelines that integrate diverse data types—including text, images, audio, and video—into enterprise AI applications. This session will cover the full journey, from designing production-ready retrieval systems to effectively using them in Enterprise use cases.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

How can an LLM effectively navigate the diverse complexities of enterprise systems and data - emails, documents, calendars, messages, tasks, tickets, line of business systems? How do you enable precise orchestration and execution of multi-step workflows while ensuring accuracy, predictability, and reliability in production? This session will explore how Dropbox Dash leverages cutting edge advances in RAG and Agents to tackle these challenges. From integrating heterogenous data sources and optimizing task orchestration by leveraging advances in powerful LLM models, we share our journey in building scalable and secure AI solutions. We will discuss how robust evaluation frameworks and architectural innovations help us ensure reliable, enterprise-ready systems.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

As the context length of LLM serving continues to increase, the inference time escalates dramatically. To strike this issue, we devise three key techniques: A fine-grained sparse attention algorithm for the pre-filling phase, equipped by a high-performance token-sparsity attention kernel using Tensor Core with CUTLASS CUTE, which delivers an 8X speedup at a 10X sparse rate without sacrificing accuracy; block-sparsity attention for the decoding phase, which further doubles the sparse rate while preserving accuracy with novel KV cache clustering; dynamic chunked pipeline parallelism, which decomposes a long sequence into chunks with adaptively updated chunk size based on an analytical cost model to balance pipeline stages, resulting in about 2X acceleration over Tensor Parallelism on eight GPUs. The above optimizations have empowered over 20X and 4X speedups for the pre-filling and decoding phases, respectively, while preserving accuracy with context length of 1 million tokens.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.

AI agents and agentic architectures offer significant potential as enterprises move from initial experiments to widespread generative AI deployments. In this session, moderated by NVIDIA’s vice president of enterprise AI and automation, AI experts will share their insights and experiences in developing and deploying generative AI and agentic solutions. They'll discuss the opportunities AI agents bring, real-world enterprise use cases, strategies for building user trust, and the challenges of scaling these technologies across organizations. Join us as we explore the future of AI in the enterprise.

Important: Near capacity, highly suggest arriving early. Attendees are let in on a first-come, first-served basis.