Skip to main content

03.Build / learn / document

Docs
& notes

Practical notes on product development, engineering, and running an AI-native one-person company.
Browse the docs

Notes from real projects

Field notes

10articles
01 Getting started

What the Owner Is Working On

A current record of the products, AI infrastructure, and GeekCode writing I am working on—including what is not finished yet.

02 engineering practice

Should You Write Tests When Vibe Coding?

Tests can be executable memory for Vibe Coding, but they can also consume context, slow regression, and freeze iteration. The real question is which evidence each risk deserves.

03 engineering practice

Engineering Stability for Vibe Coding: Specs, Architecture, and Changelogs

Turn an AI-assisted change into an admitted, implementable, verifiable, traceable, and reversible engineering milestone. Domain boundaries and testable frontend state are key parts of the larger delivery system.

04 engineering practice

When an API Gets Slow, Do Not Add Machines Yet

The first job in a slow-endpoint investigation is not “optimization.” It is finding out where the time actually goes.

05 ai perspectives

Beyond Prompt Tuning: Four Guardrails for Reliable AI Features

A feature that demos well is not necessarily ready to ship. Keep uncertainty at the boundary before asking users to rely on it.

06 independent development

Set Up Your First Environment

A practical environment-setup guide is being prepared for this chapter.

07 independent development

SaaS or Self-Hosted Services? A Practical Choice for Indie Developers

How independent developers can weigh hosted services against systems they operate themselves.

08 independent development

Why Every Developer Should Try Independent Development

A comprehensive guide to understanding independent development, its charm, and value

09 Getting started

Getting started

An introduction to GeekCode, what it builds, and how this documentation is organised.

10 ai perspectives

What Is MCP?

An introduction to how the Model Context Protocol connects LLM applications with tools and external information.