Who writes this blog
Hi. I'm a software engineer with 18 years of experience. I've worked across large-scale backend system design, data pipeline construction, cloud infrastructure operations, and most recently LLM-based AI agent development in production environments.
Currently I focus on applying AI agent workflows, LangGraph, and RAG pipelines to production systems. This blog documents real lessons learned from that work.
Technical background
- AI/LLM: LangGraph agent design, RAG pipelines, MCP server development, prompt engineering
- Backend: Large-scale services in Java/Spring Boot, Python, and Golang
- Data: MySQL performance tuning, Redis/Kafka/RabbitMQ operations, ETL pipelines
- Infrastructure: Production operations on Kubernetes, Docker, AWS, and GCP
What the blog covers
- Production AI: Agent architecture, RAG, prompt design at production scale
- Troubleshooting: MySQL deadlocks, Kubernetes OOMKilled, Redis memory issues — real incident resolution
- Dev tools: Practical usage of AI coding tools like Claude Code and Codex
- Infrastructure: Deployment, monitoring, and performance optimization guides
Writing approach
Every post on this blog is grounded in real problems encountered, code actually written, and systems actually operated. The goal is to help readers quickly determine why a problem happens, what to check first, and what the most realistic next step is — with terminal commands, config examples, and decision criteria included.
AI tools may assist with draft organization, but technical accuracy and structure are always reviewed manually before publication.
Contact
For error reports, technical discussions, or collaboration proposals, reach out at admin@hobokai.com.