MS student in Intelligent Information Systems at Carnegie Mellon's School of Computer Science, working on machine learning systems.https://yash-agarwal.org/https://yash-agarwal.org/blog/blasst/https://yash-agarwal.org/blog/blasst/BLASST block-sparse attention plus one TMEM address swap: −41.9% attention-kernel time at 95% skip and −20% TTFT at 64K with accuracy above dense — an optimization worklog on a B200.Mon, 31 Aug 2026 00:00:00 GMThttps://yash-agarwal.org/blog/dspark/https://yash-agarwal.org/blog/dspark/How speculative decoding for Gemma4 on Modular's MAX went from 30% behind vLLM under load to outside its latency/throughput curve on every dataset at concurrency 8 and 64, on one B200.Mon, 31 Aug 2026 00:00:00 GMThttps://yash-agarwal.org/blog/pinned-host-memory/https://yash-agarwal.org/blog/pinned-host-memory/9× faster than cuMemAllocHost, and faster than CUDA's own VMM API too — an optimization worklog on an 8×B200 host.Mon, 31 Aug 2026 00:00:00 GMT