AI & ML
{{ topic.name }}
No items found.
Is purpose-built for AI training at scale really better? If so, how much better?
Our engineering team set out to answer this question, which led to months of research, testing, and even our own trained AI model, all captured in our latest performance benchmarking whitepaper.
Join Distinguished Engineer Wes Brown and Product Manager Deok Filho as they pull back the curtain on the methodology, the surprises, and the hard-won optimizations that delivered up to 20% more throughput, 10x longer uptime, and 97–98% utilization.
In this session, you’ll learn:
- The hard data, charts, and benchmarks that prove an AI-first cloud outperforms industry training benchmarks
- How we measured MFU, MTTF, and ETTR at massive scale—and why those metrics matter
- What optimizations move the needle, from high-throughput tokenization to async checkpointing and automated recovery
- Actionable next steps for applying these measurement and optimization techniques to your own AI training workflows
TCO Insights for 2025
Budget pressures shouldn’t stall AI innovation. Join CoreWeave and guest Forrester as we decode the economics behind cutting-edge AI infrastructure. Leveraging Forrester’s industry research and CoreWeave’s experience powering AI infrastructure for leading AI labs and enterprises, this webcast demystifies total cost of ownership (TCO) for AI workloads. Gain a practical lens on cost drivers, from GPU generations and supporting AI infrastructure to deployment models and support. Discover how to align technical architectures with financial goals and secure executive buy-in for your next AI initiative.
In this session, you’ll learn:
- How TCO for AI infrastructure is evolving in 2025
- Hidden cost drivers (job interruptions, data egress, support)
- Scenario analysis: scaling from pilot to production
- A checklist to future-proof your AI budget
Speakers
Tracy Woo
Principal Analyst at Forrester, Guest Speaker
Tracy leads Forrester’s research on FinOps, hybrid cloud management, IT financial management (ITFM), technology business management (TBM), and industry cloud in the financial services, retail, and healthcare industries. Tracy’s broader expertise focuses on cloud cost management, cloud governance, cloud skills, cloud center of excellence (CCOE), cloud platform teams, industry cloud, hybrid cloud management, and public cloud pricing. Tracy delivers strategic guidance to Forrester’s enterprise and vendor clients by helping them get cloud costs under control and build out effective cloud organizations. During her tenure at Forrester, she has advised leading Fortune 100 and Fortune 500 companies on their cloud strategies and has been frequently cited in major news outlets including TechTarget, Business Insider, The Wall Street Journal, and Computer Weekly.
Urvashi Chowdhary
VP, Product Management, CoreWeave
Urvashi Chowdhary is the Vice President of Product Management for AI Services at CoreWeave. She brings more than a decade of experience from AWS, where she built and scaled core cloud platforms and services. Before AWS, Urvashi held roles across strategy, business development, and engineering in the cloud computing space, giving her a broad perspective on both the technical and business dimensions of emerging cloud technologies. Passionate about working with customers, she is dedicated to helping organizations adopt and thrive with next-generation AI and cloud solutions.
How CoreWeave’s stack stays AI-first from all angles
In the rapidly evolving field of artificial intelligence, leading AI labs and enterprises are constantly seeking ways to maximize the performance of their GPU clusters in order to get most out of their infrastructure and get to market faster.
We'll explore advanced strategies and techniques for optimizing GPU clusters to achieve superior speed, efficiency, and reliability—without cutting into performance.
Key takeaways
- Dive into critical aspects such as improved model loading, better checkpointing, and robust fleet management for reduced downtime.
- Gain valuable insights into best practices and emerging technologies that are shaping the future of GPU cluster management.
- An inside look into how CoreWeave quickly identifies, remediates, and tracks job and job failures to achieve an industry-best effective training time for large-scale GPU clusters.
AI is evolving fast—are your GPU clusters keeping up? Watch the webinar to learn more or contact us to get started now.
Inside CoreWeave’s NVIDIA GB200 NVL72 Clusters
Join Jacob Yundt, CoreWeave’s Director of Compute Architecture, and Harsh Singh Banwait, Director of Product, for an in-depth exploration of CoreWeave’s NVIDIA GB200 NVL72-based instances. Discover how these cutting-edge superchips power today’s most demanding AI workloads and how CoreWeave helps to maximize the NVIDIA GB200 NVL72's capabilities to deliver incredible performance and resilience for our customers.
Key takeaways
- Discover the core components of the NVIDIA GB200 NVL72 rack-scale solution on CoreWeave, including liquid-cooled, high-density racks
- Explore how CoreWeave’s purpose-built architecture, including CoreWeave Kubernetes Service (CKS) and Slurm on Kubernetes (SUNK), is optimized for high performance
- Learn how to unlock up to 4x faster training and 30x faster real-time inference with NVIDIA GB200 NVL72 clusters on CoreWeave
Be among the first to harness the power of a NVIDIA GB200 NVL72 -based instances. Register for the webinar to learn more or contact us to get started now.
Improve the observability and reliability of your AI cluster.
Ready to elevate your Kubernetes cluster management skills? Watch CoreWeave’s CTO, Peter Salanki, and SVP of Engineering, Chen Goldberg, discuss strategies to improve full-stack observability and reliability with AI fleet management.
Gain practical knowledge of building more reliable and efficient AI operations in Kubernetes.
Key takeaways
- Uncover critical components of a large-scale Kubernetes training cluster optimized for AI workloads.
- Learn how advanced fleet management techniques can enhance cluster resilience and accelerate time-to-market for AI models.
- Discover how automation can help detect, diagnose, and respond to job failures, minimizing downtime.
- Gain insights via comprehensive monitoring across all layers of your AI infrastructure stack.
Keep job interruptions to a minimum, and know why they happen when they do. Register for the webinar today.