Category: Work

2025

• Nov 21
  What I Hope to Do in the Long Term
• Oct 22
  How AI Looks Like in 2030?
• Sep 24
  We're Coming Out of Stealth Mode
• Aug 14
  The Headroom of Optimization
• Jun 29
  New Chapter in 2025
• Jun 11
  KubeCon HongKong - Build a Large Model Inference Platform for Heterogeneous Chips Based on vLLM (Keynote)
• Jun 10
  KubeCon HongKong - New Pattern for Sailing Multi-Host LLM Inference
• Apr 11
  KubeCon EU 2025 Summary
• Apr 2
  KubeCon London - Sailing Multi-Host Inference with LWS
• Mar 5
  KServe, AIBrix, and llmaz
• Feb 21
  OpenSource Commercialization With Fit2Cloud
• Feb 6
  Does DeepSeek Break CUDA Moat?
• Jan 28
  Random Thoughts of DeepSeek
• Jan 10
  What I Gained And Paid For Open Source

2024

• Aug 23
  KubeCon HongKong - Panel: Fragmentation of the Scheduling in Kubernetes and Challenges for AI/ML Workloads
• Mar 22
  KubeCon Paris - SIG-Scheduling Intro & Deep Dive

2023

• Sep 27
  KubeCon Shanghai - SIG-Scheduling Intro & Deep Dive
• Sep 27
  KubeCon Shanghai - Sailing Ray workloads with KubeRay and Kueue in Kubernetes
• Apr 19
  KubeCon Amsterdam - SIG-Scheduling Deep Dive
• Apr 18
  KubeCon Amsterdam - Building a Batch System for the Cloud with Kueue

2022

• Oct 24
  KubeCon Detroit - SIG-Scheduling Deep Dive

2021

• Nov 6
  KCD Shanghai - Panel: How OpenSource Contribution Impacts Personal Career Development