This post breaks Agent development down into three main tracks — modular design, outcome‑driven delivery, and continuous evolution — and maps them to concrete components and practices.
Technical and engineering practice notes organized in reverse chronological order.
This post outlines a layered evolution path for platform engineering — from standardization and automation to API‑driven capabilities and self‑service, emphasizing thin slices that validate end‑to‑end delivery.
DomainSpec uses a set of transferable specification artifacts to carry business context from discussion all the way through implementation and verification, reducing semantic drift and delivery misalignment.
A recap of weaving AI into a Mesh-Shaped PKM system, shifting focus from "better note-taking" to "faster task completion" and building feedback loops that improve execution.
Grounded in real usage scenarios, this post examines how AI boosts efficiency and triggers replacement anxiety, and argues for treating AI as an amplifier instead of a substitute.
Traces how LLM apps evolve from basic prompts to vector search, caching and orchestration layers, and shows how to build a local stack with Ollama and LangChain.
Extends diagrams from an expression tool to a thinking tool, with a four-step method – reproduce, diverge, converge, recombine – to build reusable cognitive structures in learning and work.
A look at the real frictions in cloud-native adoption – tool overload, collaboration gaps and hard-to-measure value – and an engineering mindset centered on API-first and data-driven decisions.
Practical notes on using DevContainer for multi-language projects – defining, building, connecting and sharing containers to reduce local drift and version conflicts.
A practice-driven framing of how Agile and Lean overlap in software teams – goals, methods and organizational evolution – with a focus on what is actually executable.