Codeaholicguy
I’m Hoang Nguyen, VP of Engineering at ShopBack, with more than a decade of building software and leading teams. I write about software engineering, leadership, and AI coding agents, and send each new post to my newsletter.
Latest post
Agentic engineering runs on boring foundationsHow multi-tenancy, service consolidation, governed data access, and verification workflows became the foundation for agentic engineering.
· 12 min read
Previous posts
- AI coding agent session compaction with Jev
How AI DevKit compacts coding agent sessions with Jev typed decisions so manager, executor, and reviewer agents hand off state instead of raw transcripts.
· 8 min read
- MCP vs Skills
Will skills replace MCP? Why agent skills are the instruction layer and MCP is the integration layer, where the two overlap, and how they work together.
· 5 min read
- How I run AI coding agents as a team with AI DevKit
How I run AI coding agents as a team with AI DevKit: a manager agent delegating to Codex and Claude Code executors, Telegram, skills, and shared memory.
· 11 min read
- I read every Claude Code hook so you don’t have to
A practical guide to all 27 Claude Code hook events: how hooks work, which are worth using, like PreToolUse and SessionStart, and pitfalls to avoid.
· 5 min read
- A practical roadmap to learn agentic engineering
A staged roadmap to learn agentic engineering: master one coding agent, context engineering, verification, and workflows before MCP and multi-agent work.
· 9 min read
- More agents, same human brain
Running Claude Code, Codex, and Gemini CLI in parallel showed me the bottleneck is human attention, plus my agent orchestration experiments.
· 12 min read