AI coding for teams

Claude Code training and setup for development teams

I help development teams work with Claude Code, Codex and Cursor. I set the tools up in your repositories, show your developers how I use them, and help you agree on rules for review, testing and what goes live.

I'm all in on AI and write most of my own code this way. It makes me fast, and I stay careful about the code and what goes live. I work with teams that are just starting and with teams whose developers already use AI but get uneven results.

I usually reply within one working day.

Vincent Schmalbach

What I help teams with

Claude Code, Codex, Cursor and Copilot, set up for your code and your way of working.

Setup in your repositories

Project instructions, permissions, sandbox settings, API keys or subscriptions, and the commands the AI is allowed to run. The tools learn your code and can't touch what they shouldn't.

Claude Code training for developers

I work through tasks from your own project with your developers and show how I plan a change, let the AI write it, and check the result.

Claude Code for teams and enterprise

Shared settings, which plan or API setup makes sense, how to keep costs visible, and how to handle access to production data and secrets.

Review and testing rules

AI writes code faster than people can read it. We agree on what every change needs before it is merged: tests, a human review, and small pull requests.

Cursor, Copilot and Codex

Not every developer likes the same tool. I have used all of them and can say where each one is strong.

AI pair programming in daily work

Debugging, refactoring, writing tests, reading an unfamiliar codebase. I show on your own code where AI saves the most time in these tasks.

How I work

You don't need to have everything figured out before we talk. An idea, access to the repo, a screen recording or a list of problems is enough. I look at what is there, tell you what I would do and what it costs, and then I build it.

I use Claude Code, Codex and my own agent workflows for reading code, planning changes, writing tests, refactoring and debugging. It is not vibe coding. It is senior engineering with a much faster loop.

AI-first, every day

I use Claude Code, Codex and my own agent workflows for almost all of my client work. What I show your team is how I work myself, including the checks I never skip.

I stay in control

AI writes a lot of my code. I still make every technical decision, read every change before it ships, and test it. Fast, but meticulous: the speed is only worth something if what goes live is right.

The way you want to work

Some clients want to look at every change before it goes live, others want me to deploy. Some want a weekly call, most prefer written updates. I work well asynchronously, and I still ask the important questions when they come up.

I think like an owner

I still run my own online products. So when I build or fix yours, I think about traffic, conversion, revenue and support as much as about the code, and I tell you when something is not worth building.

Examples of my work

Where my AI coding experience comes from.

My own work

AI-first on every client project

I plan changes, write tests, refactor and debug with Claude Code, Codex and my own agent workflows. I make the technical decisions and read and test every change before it ships.

Most-read on my blog

Guides on Claude Code, Codex and Cursor

The most-read posts on my site explain how Claude Code uses an API key instead of a subscription, how the Codex sandbox and CLI flags work, and how Copilot, Cursor, Cody, Supermaven and Aider compare.

My own tools

Go CLIs made for AI coding agents

I wrote rungrad, an open-source Go framework for CLIs that AI agents and scripts can use safely, and keep a public list of CLIs that work well with AI coding agents.

What clients say

"Vincent works very well asynchronously: communication stays focused, there is little unnecessary back-and-forth, and he still raises the important questions when needed."

Aidas K., Bizdras OU

"We especially value how quickly he gets into mature projects and finds solutions that fit the existing state of the codebase."

Florian Brunner, designundzwanzig OG

"He also turned customer discussions into clear technical requirements and documented his work so the rest of the team could build on it."

Yash Chandra, Academy of Mine

Questions

Is this a workshop or ongoing help?

Either. Some teams want a focused session on their own code. Others want me to stay for a while, review how the tools are used and adjust the setup.

Claude Code, Cursor or Copilot?

It depends on your team and your code. I use Claude Code and Codex most, and I'll tell you where Cursor or Copilot fits better.

Is our code safe with these tools?

Part of the setup is deciding what the tools can read, which commands they can run, and which data never goes to a model provider. We choose plans and providers whose terms fit your requirements.

Our developers are sceptical. Does that matter?

I show them the results on their own code and the checks I use. Then they decide for themselves.

Do you also write code for us?

Yes. I can train the team and also take on development work. Many teams start with one and add the other.

How do you charge?

Fixed price for a job with a clear outcome, a flat monthly rate for ongoing work, or hourly. Describe what you need and I'll suggest one.

Get in touch

How big the team is, which tools you use today, and what you want to get out of AI coding. A few sentences are enough.

If you would rather talk first, book a 30-minute call or email contact@vincentschmalbach.com.

I usually reply within one working day.