資深後端工程師，持續研究分散式系統、可靠性工程、基礎架構與 AI 協作開發。
Backend Engineer focused on building and understanding reliable software systems.
My work and experiments mainly revolve around backend architecture, distributed systems, performance, infrastructure, and AI-assisted software engineering.
I care not only about whether software works, but also about why it works, how it fails, how we verify it, and how we improve the engineering process around it.
- C# / .NET / ASP.NET Core
- REST APIs and backend services
- Domain-Driven Design
- Clean Architecture
- Distributed Systems
- Microservices
- MySQL / MariaDB / SQL Server
- Redis / MongoDB
- RabbitMQ / Kafka
- Concurrency and data consistency
- Docker / Kubernetes
- NGINX
- Google Cloud
- CI/CD
- Observability
- Grafana
- k6
- Performance & load testing
I build and maintain Kubernetes environments for hands-on experimentation with:
- cluster architecture
- networking
- ingress / gateway
- high availability
- resource management
- observability
- performance testing
- failure investigation
I use these environments not only to learn Kubernetes, but also to understand how application, infrastructure, network, and data-layer behavior interact as a system.
I'm actively exploring how AI coding agents can participate in real engineering workflows.
My focus is not simply generating code with AI.
I'm interested in engineering workflows where AI agents can assist with:
- implementation
- code review
- investigation
- testing
- documentation
- architecture analysis
while keeping verification, evidence, review, and human decision gates explicit.
The goal is to understand where AI can improve engineering productivity — and where engineers still need to understand the underlying system and make the final judgment.
- Distributed systems
- System design
- Software architecture
- Reliability engineering
- Concurrency and race conditions
- Performance engineering
- Kubernetes internals and networking
- AI coding agents and engineering workflows
- Human–AI collaboration in software development
A large part of this account is my long-term engineering notebook.
Some repositories are experiments, learning records, prototypes, or references accumulated over different stages of my engineering journey.
I prefer keeping that history rather than presenting every repository as a polished production project.
Understand the system.
Verify the evidence.
Then make the engineering decision.