6+ years building scalable backend systems for SLG games. Go expert. Currently focused on AI-augmented development workflows and distributed systems tooling.
I'm a backend engineer specializing in high-concurrency game server architecture, with additional hands-on experience in Unity frontend development. I've spent the last 6+ years deep in the trenches of SLG (Simulation/Strategy Game) development β building both the backend infrastructure that handles thousands of concurrent users, cross-server synchronization, and real-time state management, as well as the Unity-based game clients that players actually interact with.
This full-stack game development background gives me a unique perspective β I don't just design APIs and infrastructure; I understand how they're consumed on the client side.
Currently exploring the intersection of AI and developer tooling β specifically how LLMs and MCP can revolutionize the way teams build and operate backend systems.
When building internal tools and dashboards, I also work on the web frontend β typically with Vue, turning backend services into usable interfaces for the team.
Planning to open source components extracted from production systems. Currently working on:
- [WIP: kratos-extensions] β opinionated extensions and utilities for Kratos framework
- [WIP: go-middleware-kit] β battle-tested middleware for Go services (rate limiting, circuit breaking, tracing)
Fun fact: I've starred the entire OpenClaw ecosystem on GitHub. Lobster enthusiasm.
- Cross-Server Communication Annotations β A code annotation system that makes cross-server calls feel like local calls. Significantly reduced cross-server development complexity for my team.
- Dify + MCP Integration β AI-assisted development toolchain built on top of Dify and MCP protocol, bringing AI capabilities directly into the development workflow.
- Hermes Agent β Currently researching multi-agent orchestration frameworks and their applications in game server operations and developer workflows.
- Human-AI Collaboration Platforms β Exploring platforms like Multica that bridge human workflows with AI agents, evaluating how they could fit into game development and operations.
Core: Go (expert) Β· Python Β· Lua Β· C++
Game Dev: Unity (C#) Β· Game Client Β· SLG Mechanics
Web Frontend: Vue.js Β· Internal Tools & Dashboards
Infra: Kubernetes Β· Docker Β· Redis Β· MongoDB Β· MySQL Β· gRPC
Architecture: Microservices Β· DDD Β· Distributed Systems Β· Event-Driven
AI: Dify Β· MCP Protocol Β· LLM Integration
| Area | What I Care About |
|---|---|
| Performance | Latency, throughput, memory efficiency β especially under load |
| Developer Experience | Tooling, automation, AI-assisted workflows |
| Reliability | Observability, tracing, fault tolerance |
| Architecture | Clean boundaries, maintainable microservices |
Running on OpenClaw β the AI assistant that actually gets things done. Yes, I'm a lobster.
- π§ [email protected]
- π https://muzig.github.io/starrypen
- π¬ Zhihu: ει’η΄θ΅°
Last updated: 2026-04