NVIDIA AI Podcast How AI Data Platforms Are Shaping the Future of Enterprise Storage - Ep. 281
13 snips
Nov 18, 2025 Jacob Liberman, Director of Enterprise Product Management at NVIDIA, dives into how GPU-accelerated storage transforms enterprise data management. He explains the challenges of integrating AI agents in production and the critical role of secure data access. Liberman highlights the importance of making unstructured data AI-ready, discusses the security risks of data copies, and illustrates how NVIDIA's AI Data Platform enables continuous processing and efficient workflows. Anticipate the evolution of storage solutions through innovative partnerships and real-world implementations.
AI Snips
Chapters
Transcript
Episode notes
Data Access Is The Core Enterprise Challenge
- Enterprises struggle to deploy AI agents mainly because they lack secure, recent access to their data.
- Jacob Liberman emphasizes that all AI relies on accurate, up-to-date data regardless of model or use case.
Pipeline For Making Unstructured Data AI Ready
- Making unstructured enterprise data AI ready requires extraction, chunking, metadata enrichment, embedding, and indexing.
- Jacob Liberman describes this pipeline as necessary for retrieval-augmented generation at scale.
Data Velocity Forces Continuous Processing
- Data velocity (growth plus change) forces continuous reprocessing to keep AI representations current.
- If you can't detect changed items, organizations end up re-indexing everything, wasting resources.