Notes
What I learn along the way: building with AI, agents, and retrieval.
2026
- A Playbook for Building Real Apps with AIGate-Driven AI Development: five phases, five gates, behavior evidence: how a solution architect ships a real app with AI doing all the engineering.
- Four Hours, Fourteen Architectural MistakesA follow-up on working with AI engineers: why the hardest skill in AI-assisted development is learning how to challenge the AI.
- The Team I Never Had BeforeA reflection on building with AI as a real engineering team, from design through coding to deployment.
- How Domain Experts Can Build AI Agent SkillsA practical guide for turning expert workflows into reusable AI agent skills, using a medical monitor review workflow as an example.
- AI Agents Need Retrieval and SkillsA practical framework for building AI agents that retrieve trusted knowledge, apply reusable skills, or combine both.
- Three-Layer Thinking for Building Knowledge AgentsHow surface, essence, and philosophy can improve LLM use and help design better knowledge agents.
- Is RAG Dead?A practical reflection on why the concept behind RAG still matters, even as naive RAG becomes insufficient for domain-specific AI agents.
- Beyond Vibe Coding: From Simple Prompts to Spec-Driven DevelopmentWhy fast AI coding breaks as systems grow, and how spec-driven development helps AI-generated software scale.
- My Last Line of CodeA personal reflection on writing less code, designing more intent, and moving toward structured natural language.
2025
- Selecting the Best Embedding Method for Limited Domain KnowledgeA comparison of chunk-based and query-focused embedding strategies for private-domain agents.
2024
- Enhancing Query Retrieval Precision Through Optimized Embedding Text SelectionResearch notes on how embedding construction affects semantic search precision.