beixu@home:~$ cat /work/ai-infrastructure.md

ai infrastructure

Cloud-native AI infrastructure, resource scheduling, GPU management, and platform architecture.

What this is

Infrastructure work behind applied AI: cloud platforms, container systems, schedulers, GPU management, and deployment surfaces that let models and agentic workloads run reliably.

Tencent Cloud

Public cloud container platform architecture, AI infrastructure, resource scheduling, and GPU virtualization systems.

TalkingData

Cloud-native big data platform, private cloud cluster management, task scheduling, microservices, and multi-cloud architecture.

Hewlett-Packard

Private cloud platform architecture, core modules, resource management, and infrastructure development.

GPU pod tooling

vLLM-oriented deployment and GPU pod management experiments for agentic workloads.

Core thread

This is the base layer under the rest of the work: before agents, robotics, or clinical AI can be useful, they need reliable compute, orchestration, observability, and deployment paths.