Xiaozhe YaoXiaozhe Yao's personal websitehttps://about.yao.sh/Build Neural Network From Scratchhttps://about.yao.sh/posts/build-nn/https://about.yao.sh/posts/build-nn/Building Neural Network from Scratch using Python and Numpy onlyThu, 01 Apr 2021 00:00:00 GMTVision Debugging Benchmark @ DataPerfhttps://about.yao.sh/posts/dataperf/https://about.yao.sh/posts/dataperf/Lessons learned from the DataPerf Vision Debugging BenchmarkWed, 01 Feb 2023 00:00:00 GMTDeltaZip: Serve Multiple Full-Model-Tuned LLMshttps://about.yao.sh/posts/deltazip/https://about.yao.sh/posts/deltazip/DeltaZip: Efficient Serving of Multiple Full-Model-Tuned LLMsWed, 23 Apr 2025 00:00:00 GMTHow SwissAI Uses OpenTela for Scalable LLM Servinghttps://about.yao.sh/posts/opentela-swissai/https://about.yao.sh/posts/opentela-swissai/How SwissAI Uses OpenTela for Scalable LLM ServingFri, 06 Mar 2026 00:00:00 GMTReflection on Building Swiss AI Servinghttps://about.yao.sh/posts/swissai-serving/https://about.yao.sh/posts/swissai-serving/Reflection on Building Swiss AI ServingSat, 30 Aug 2025 00:00:00 GMTWhy Delta Compression Works - Information Theoretic Perspectivehttps://about.yao.sh/posts/why-delta-works/https://about.yao.sh/posts/why-delta-works/Understanding the effectiveness of Delta Compression in LLMsMon, 21 Jul 2025 00:00:00 GMT