Agents 6 min
Task Decomposition for AI Coding Agents: How to Break Down Complex Features
How to split a complex feature into agent-sized sub-tasks that execute reliably — with concrete patterns and scope rules that prevent drift and rework.
Josh
[ Index ]
Articles
Page 03 / 04
Fundamentals 5 min
How LLMs Actually Work: The Mental Model Every AI Engineer Needs
The practical model behind next-token prediction, tokens, context windows, and sampling — the fundamentals every other post on this site assumes you already have.
Josh
Tooling 7 min
Upgrading Dependencies with AI: A Systematic Migration Workflow
How to use AI coding assistants to navigate major library upgrades — feeding changelogs, mapping the breaking change surface, and validating the result.
Josh
Agents 7 min
Steering a Drifting AI Agent: How to Correct Course Mid-Task
Practical techniques for recognizing when an AI coding agent has gone off track and redirecting it effectively — before it digs you deeper into the wrong solution.
Josh
Tooling 7 min
Using AI to Explore Unfamiliar Codebases
How to use AI coding assistants as a guide when navigating code you didn't write — from getting an architectural overview to tracing execution paths.
Josh
Tooling 6 min
Refactoring with AI: A Step-by-Step Strategy for Modernizing Legacy Code
How to use AI coding assistants for safe, incremental refactoring — scoping the work, writing safety tests, reviewing generated diffs, and avoiding common pitfalls.
Josh
Agents 7 min
MCP Servers: Giving Your AI Coding Agent Real Tools
How to extend your AI coding assistant with Model Context Protocol servers — connecting it to your database, APIs, and custom tools in about ten minutes.
Josh
Production 7 min
Writing Tests with AI: Getting Edge Cases, Not Just Coverage
How to prompt AI coding assistants to generate tests that actually catch bugs — edge cases, failure modes, and boundary conditions, not just happy-path boilerplate.
Josh
Tooling 7 min
Managing Context in Long AI Coding Sessions
How to keep AI coding assistants accurate across a long feature branch — when to trim, reset, and summarize context so the model stays focused.
Josh