I build practical infrastructure automation: from Kubernetes operations and observability to AI agents that can diagnose and manage production systems.
- Building AI-native operations workflows for Kubernetes and cloud infrastructure.
- Turning runbooks into reusable agents, MCP servers, and automation skills.
- Improving reliability through observability, GitOps, and repeatable tooling.
- Contributing fixes and documentation back to the open-source projects I use.
| Project | What it does | Built with |
|---|---|---|
| Agentic Infra | Turns natural language into Kubernetes deployment, diagnosis, and Day-2 operations through specialized agents and 87+ MCP tools. | Python, FastAPI, Next.js, Kubernetes, Argo CD |
| Stock Intraday Trading | An agent skill and CLI for A-share screening, multi-timeframe analysis, intraday trading research, and cost-aware backtesting. | Python, TypeScript |
| Xray MCP | Generates and validates Xray-core configurations through natural-language MCP tools. | TypeScript, MCP, Xray-core |
| dsflow | Reusable SRE inspection workflows for Kubernetes, Ceph, compute clusters, and on-call reporting. | Python, Shell, Kubernetes |
Kubernetes platform engineering ── SRE automation ── Observability
AI agents + MCP ── GitOps ── Open source
If you're working on cloud-native operations, infrastructure agents, or useful developer tooling, feel free to start a conversation.




