As a Software Engineering Technical Leader at Cisco, Ram Gandikota’s core focus is AI Infrastructure Solutions and Performance Benchmarking for modern AI workloads. He evaluates, validates, and characterizes the network, storage, and compute infrastructure that supports distributed training and inference, helping maximize accelerator efficiency and achieve consistent system performance. Ram specializes in end-to-end benchmarking across RoCEv2, Collective communications, and AI workloads, working with industry organizations and representing Cisco at MLCommons through MLPerf benchmarking efforts. He works with partners, customers, and stakeholders across Cisco to drive performance and enable predictable scaling. Through his technical blog posts, he shares practical insights into AI workload performance and the infrastructure required to scale across accelerator ecosystems.