Event details
NVIDIA HGX B300 on CoreWeave: What Changes for Agentic AI at Scale
Get the infrastructure you need to scale agentic AI
Agentic AI workloads introduce long-context processing, chained execution, and unpredictable utilization patterns that strain traditional cloud infrastructure. As enterprise adoption accelerates, platform readiness—not just GPU access—determines performance, reliability, and cost efficiency at scale.
In this 30-minute product briefing, CoreWeave explains what’s new with NVIDIA HGX B300 and why CoreWeave Cloud is best suited to unlock NVIDIA HGX B300 performance at scale.
In this webinar, we’ll cover:
- Why agentic AI changes memory, bandwidth, and latency requirements
- What’s new with NVIDIA HGX B300 on CoreWeave
- What materially changes vs previous-generation systems
- How CoreWeave Cloud unlocks full B300 performance at multi-node scale
- Benchmark results and methodology highlights
- Which teams should evaluate adoption now
Speakers
Jacob Yundt is Senior Director of Compute Architecture at CoreWeave, where he leads teams spanning hardware, firmware, systems engineering, compute fabrics, and platform software. Since joining in April 2023, he has helped drive the launch of industry-leading AI infrastructure, including being first to launch GB200 and GB300.
Harsh heads CoreWeave's Product team, focusing on crafting intuitive yet impactful developer experiences. Over the better part of the last decade, he has contributed to building infrastructure products at both established Fortune 500 companies and dynamic startups.
