Practical guides on the tools developers use every day — JSON, regex, JWT, cron, and more.
18 posts matching · page 1 of 2
A practical, no-hype orientation to working effectively with AI coding agents day to day, covering delegation, review, and where to hold the line.
Categories of engineering work developers underuse AI for today, despite being a strong match for how these tools actually perform best.
A candid list of the requests developers commonly make of AI coding tools that reliably produce worse outcomes than doing the work directly.
An honest look at how well AI agents can diagnose and fix failing CI pipelines, and the categories of failure where they tend to get it wrong.
How AI coding agents generate test cases and use test results as feedback, and where their test-writing instincts need a second look.
A git workflow for working with AI coding agents that keeps history readable and makes it easy to undo a bad run without losing good work.
How to select the right context for an AI coding agent so it has what it needs without hitting context-window limits or drowning in noise.
Concrete techniques for getting an AI coding tool to match your project's conventions instead of defaulting to generic, textbook-style code.
The categories of engineering work where AI coding agents most often produce confidently wrong results, and why those tasks resist automation.
A practical rundown of the kinds of engineering tasks where AI coding agents consistently deliver reliable, time-saving results.
Guardrails worth setting up before giving an AI agent write access to your repository, from sandboxing to branch protection.
Explains the concrete techniques AI coding agents use to locate bugs, from reproducing failures to reading stack traces and bisecting changes.