https://losslessnetwork.com/podcasts AI networking for network engineers — distilled into 10–15 minute conversations. RoCE, RDMA, lossless Ethernet, the fabrics behind the AI models you keep hearing about, translated into the language of someone who already runs a network. en-us © 2026 Nagarjun Velmurugan. Licensed CC BY 4.0. Nagarjun Velmurugan AI networking for network engineers — distilled into 10–15 minute conversations. RoCE, RDMA, lossless Ethernet, the fabrics behind the AI models you keep hearing about, translated into the language of someone who already runs a network. Nagarjun Velmurugan arjun@losslessnetwork.com false https://losslessnetwork.com/img/lossless-social-card.jpg https://losslessnetwork.com/podcasts In this first conversation we cover the three things every network engineer needs to internalize before they look at an AI cluster: (1) why tail latency on one link kills the whole job, (2) why "lossy" Ethernet isn't survivable, and (3) why oversubscription — the design lever you've relied on for two decades — becomes the enemy here. Aimed at network engineers who've never racked a GPU server and want to. A 12-minute primer on why your DC playbook doesn't carry over to an AI fabric — and what changes when 8,000 GPUs gradient-AllReduce every 2 seconds. Nagarjun Velmurugan 1 full 763 https://losslessnetwork.com/podcast/episodes/ep01-why-ai-networks-are-different Sat, 23 May 2026 12:00:00 GMT https://losslessnetwork.com/podcasts#ep01-why-ai-networks-are-different