Claude Code for Engineers — From Chat to Coworker
Most engineers use Claude like a smarter search box. Here is the short path to using it like a teammate that actually ships work on your machine.
- #claude-code
- #terminal
- #engineering
I run engineering at a US startup, and the single biggest gap I see between people who get value from Claude and people who do not is this: most engineers are still typing questions into a chat box and copy-pasting answers back into their editor. That is the slowest possible way to use it.
Claude Code closes the loop. It reads your files, runs your commands, and edits code in place — on your machine, in your repo. The jump from "smarter Google" to "coworker who does the work" is mostly about two habits.
Habit 1: write a CLAUDE.md
Drop a CLAUDE.md in a folder and Claude picks it up on its own whenever you start working there — no copy-pasting context into the prompt each time. It is the difference between onboarding a stranger every morning and working with someone who already knows your stack, your conventions, and the three things you never want them to touch.
Keep it short and specific. Mine usually has four sections:
- Stack — languages, framework, package manager, how to run tests.
- Conventions — naming, error handling, the patterns you actually use.
- Boundaries — what not to change without asking (migrations, auth, billing).
- Commands — the exact build, lint, and test commands.
Habit 2: give it the tools, then let it loop
Once context is solid, stop micromanaging. Let Claude read, plan, edit, and run the tests itself, then react to the result. A typical flow looks like:
# point it at the work, not the keystrokes
claude "add a guides-index.json build step, run the tests, fix until green"
The work that used to be twenty small instructions becomes one clear outcome plus the freedom to iterate. Anthropic's own guidance on context engineering makes the same point: give the model the right context and a tight feedback loop, and quality compounds.
What to do this week
Pick your most active repo. Write a one-page CLAUDE.md. Then hand Claude a real task end to end — not a question, a task — and watch where it gets stuck. Those stuck points are exactly what belongs in your CLAUDE.md for next time. That is the whole game: every session makes the next one sharper.
Sources
Keep reading
More guides like this one.
A Claude Code Plugin Adds No New Powers. It Adds Everyone Else.
Everything a plugin does, a loose .claude/ folder already does. What packaging buys you is distribution — and it costs you your command names. Here is the manifest, the install commands, and the rule for when that trade is worth making.
Four Places to Run Claude Code — and What Each One Silently Drops
A cloud session clones your repo into a fresh Anthropic VM — so your personal skills, agents and every MCP server you added with `claude mcp add` are simply not there. Surface choice isn't a UI preference, it's a capability choice. Here's the decision rule by task shape, plus the commit that makes your setup portable.
Your Weekly Setup Review Needs a Delete Column
Adding your 30th skill can quietly make the other 29 harder to find: the skill listing gets ~1% of the context window, and Claude Code drops descriptions from the skills you invoke least. Here is the Friday ritual, the routing table for each lesson, and the subtraction pass everyone skips.