You can now ask questions directly within our broadcasts using our new AskAI feature.

Designing a production-ready neocloud AI factory requires more than accelerated compute. Performance depends on bringing compute, storage, high-performance networking, orchestration, observability, and AI-native operations together as a coordinated system, with the network playing a critical role in ensuring data and workloads can move efficiently at scale. This episode will examine how networking choices interact with compute and storage, where infrastructure dependencies can create hidden bottlenecks, and how companies can build architectures that deliver predictable performance, maximize GPU utilization, and scale efficiently as AI demand grows. Key discussion points include:

• Designing high-performance network architectures for the neocloud AI factory
• Identifying infrastructure and data-movement bottlenecks that impact GPU utilisation
• Testing and validating networking, compute, and storage before production
• Building a repeatable infrastructure blueprint for predictable performance at scale