InTowards AIbyRinit Jain·Sep 11Inside a Modern AI Inference PlatformHow GPUs, model servers, LLM-D and Kubernetes work togetherA response icon1A response icon1
InTowards AIbyRinit Jain·Sep 11LLM-D ExplainedHow modern AI infrastructure routes, schedules and scales LLM inferenceA response icon1A response icon1
InTowards AIbyRinit Jain·Sep 9Kubernetes for LLM Inference: How AI Workloads Run Across a GPU ClusterHow AI workloads actually run across a GPU clusterA response icon3A response icon3
InTowards AIbyRinit Jain·Sep 9Why Traditional Load Balancing Breaks for LLMsBuilding an LLM-aware routerA response icon1A response icon1
InTowards AIbyRinit Jain·Sep 9Why Is Your LLM Recomputing the Same Prompt 1,000 Times a Day?Prefix caching, radix trees and block hashing explained
InTowards AIbyRinit Jain·Sep 7How Does One GPU Serve Hundreds of Users at the Same Time?Inside an LLM inference server
InTowards AIbyRinit Jain·Sep 7The KV Cache Explained: Why Long Conversations Get ExpensiveHow LLMs remember context without recomputing everything
InTowards AIbyRinit Jain·Sep 6What Do You Do With a Model That’s Too Big for Your GPU?Quantization, Sharding and Parallelism ExplainedA response icon1A response icon1
InTowards AIbyRinit Jain·Sep 5What really happens when you click ‘Send’ on ChatGPTA journey through modern AI Infrastructure — from one GPU to a Kubernetes clusterA response icon3A response icon3
Rinit Jain·May 15, 2025The Craziest 36 Hours of My Life: A Tale of Research, Exams, and Dance“Balancing research presentations, final exams, and a national dance competition — and somehow surviving it all.”A response icon2A response icon2