https://sakshityagi.hashnode.dev 2026-11-14T22:00:59.877Z always 1.0 https://sakshityagi.hashnode.dev/one-agent-or-five-measure-before-you-multiply 2026-11-10T13:44:57.236Z daily 0.8 https://sakshityagi.hashnode.dev/the-tool-timed-out-should-the-agent-retry 2026-11-10T13:39:21.089Z daily 0.8 https://sakshityagi.hashnode.dev/what-should-an-agent-remember 2026-11-10T13:36:00.783Z daily 0.8 https://sakshityagi.hashnode.dev/the-agent-said-done-nothing-changed-agents-that-survive-production 2026-10-24T19:34:46.909Z daily 0.8 https://sakshityagi.hashnode.dev/yesterday-it-fit-dividing-a-70b-model-across-64-gpus 2026-10-17T18:26:36.800Z daily 0.8 https://sakshityagi.hashnode.dev/self-host-vs-api-for-llms-actually-costed 2026-07-29T15:54:28.853Z daily 0.8 https://sakshityagi.hashnode.dev/making-llm-decode-cheaper-quantization-and-more 2026-07-29T15:49:26.620Z daily 0.8 https://sakshityagi.hashnode.dev/batching-and-the-idle-gpu-serving-llms-faster 2026-07-29T15:46:17.650Z daily 0.8 https://sakshityagi.hashnode.dev/the-inference-wall-where-llm-latency-and-cost-go 2026-07-29T15:34:08.865Z daily 0.8 https://sakshityagi.hashnode.dev/beyond-1d-data-parallelism-zero-tp-pp-and-cp 2026-07-19T06:45:47.089Z daily 0.8 https://sakshityagi.hashnode.dev/scaling-the-batch-and-the-cluster-gradient-accumulation-and-1d-data-parallelism 2026-07-19T06:50:44.171Z daily 0.8 https://sakshityagi.hashnode.dev/activation-recomputation-and-the-hfu-mfu-trap 2026-07-16T06:13:11.573Z daily 0.8 https://sakshityagi.hashnode.dev/the-memory-wall-where-gpu-memory-actually-goes-in-llm-training 2026-07-14T11:21:15.126Z daily 0.8 https://sakshityagi.hashnode.dev/series/scaling-llm-training 2026-11-14T22:00:59.877Z weekly 0.6 https://sakshityagi.hashnode.dev/series/serving-llms-in-production 2026-11-14T22:00:59.877Z weekly 0.6 https://sakshityagi.hashnode.dev/series/agents-that-survive-production 2026-11-14T22:00:59.877Z weekly 0.6 https://sakshityagi.hashnode.dev/archive 2026-11-14T22:00:59.877Z daily 0.5