AI Infrastructure & Operations
Validates foundational knowledge for deploying and operating the infrastructure behind modern AI workloads.
- GPU cluster architecture — node design, GPU topology, scheduling, and capacity planning for multi-GPU systems.
- High-speed fabric & networking — InfiniBand / RoCE concepts, NVLink, and low-latency interconnects for scale-out training.
- Accelerated computing workloads — training vs. inference profiles, the CUDA ecosystem, and monitoring GPU utilization.