Skip to main contentSkip to navigation

All training topics

Claude Code

Claude Code, taught on your team's own codebase

Claude Code is Anthropic's agentic coding tool. You run it in the terminal, point it at a project, and it reads files, runs commands, edits code and checks that the change works. It behaves less like a chatbot and more like a colleague with a keyboard.

Our Claude Code course is built for teams that have tried it alone and want a shared way of working. We use your repositories and your tickets, in Hebrew or English, so what people learn is already the real job.

What changes once a team has Claude Code

With a chat window you paste a snippet and paste the answer back. With Claude Code you hand over a whole task: fix this bug, add tests to this module, move the old callers to the new interface. It plans, edits several files, runs the test suite and tells you what happened.

What shifts is the division of labor. Developers spend less time typing and more time deciding and reviewing, and the chores nobody wanted to touch stop being untouchable. That needs new habits, and the habits are what the course teaches.

Who it suits, and where people start

The core audience is R&D leaders, team leads and developers who already use AI inside their editor and want to work a level up. You do not need prior Claude Code experience. You do need to know your project and how your team gets a change into production.

A growing share of attendees are not engineers. Analysts, operations people and product managers have realized the terminal is a way to automate real work, from reshaping messy files to scripts that give back hours every week. They start from one annoying process they know well.

What a session looks like

The workshop is hands-on from the first minutes. We open Claude Code on a real project of yours and learn to give instructions that produce results, to write a CLAUDE.md file that teaches it the team's rules, and to control what it may do alone and what needs your approval.

Then we move to whole tasks: a bug from your queue, tests for legacy code, a review of a change. Everyone works on their own material while we walk the room. Delivery is in Hebrew or English, and the commands stay in English, as they will at your desk.

What you leave with, and where teams stall without it

Teams that start alone tend to stall in the same places. Everyone prompts differently, nobody knows how far to trust the output, and changes that look fine in the editor break in production. With no shared rules, the benefit stays with a few enthusiasts.

You leave with a project instruction file, a review routine for code written with the tool, and a list of tasks already proven on your own repositories. The rest you practice in the full course, at your own pace.

Take it further

The full course takes you from the first run to everyday work with Claude Code. Sign-in is required to enter.