Your team learns agentic engineering in the coding agent it already uses.
We have delivered every training for engineering, product and leadership in every major coding agent: Claude Code, Codex, GitHub Copilot, Gemini CLI and Cursor. Your team trains in the tool it already uses, and a team without a coding agent yet gets one from us that fits its needs.
- Format
- Two days, remote or on site, in German or English
- Group
- 8 to 12 participants
- Price
-
EUR 1,500 per participant
Up to 12 people attend, and you pay for 10 at most (EUR 15,000)
Net, plus VAT. On site, plus travel at cost.
- Who delivers it
- AboutOne of our 70+ trainers
- Tailored
- Adapted to your team and stack in a preparation call
- Mentoring
- As needed after the training, EUR 250 per hour, typically EUR 2,000 to 12,000 in total
What changes per agent
The practice and the agenda stay the same. Per agent, the instruction files, the extension points and the approval settings change, and we set those up with your team on day one.
-
Claude Code
The agent in our hosted training environment. Every other agent runs on your team's machines and licences. Rules live in
Claude Code training for engineering teamsCLAUDE.md. Skills, subagents, hooks and MCP live under.claude, and permission modes run from manual to auto. -
Codex
Runs in the Codex CLI, IDE extension or desktop app, on your ChatGPT workspace or an API key. Rules live in
Codex training for engineering teamsAGENTS.md, custom agents are TOML files in.codex/agents, hooks and MCP sit inconfig.toml, and sandbox and approvals are two separate settings. -
GitHub Copilot
Runs on your team's Copilot seats in Copilot CLI or VS Code. JetBrains teams use the CLI. Rules live in
GitHub Copilot training for engineering teams.github/copilot-instructions.md, path-specific.instructions.mdfiles orAGENTS.md, custom agents in.github/agents, and Copilot CLI sets approvals per tool. -
Gemini CLI
Runs in your Google Cloud, on your machines or Cloud Workstations. Rules live in
Gemini CLI training for engineering teamsGEMINI.md, MCP servers and hooks in.gemini/settings.json, subagents in.gemini/agents, and the sandbox runs in Docker, Podman, gVisor or Seatbelt. -
Cursor
Runs in the Cursor desktop app or Cursor CLI, on your Cursor licences. Project rules are
Cursor training for engineering teams.mdcfiles in.cursor/rulesorAGENTS.md, hooks sit in.cursor/hooks.json, and run modes come with a sandbox. Skills and subagents from Claude Code load as they are.
Engineering, product and leadership trainings in any of these coding agents
Choose the training by who's in the room. The agent is whichever one your team already uses.
-
Your whole team working with coding agents the same way.
- Software Engineers
- QA
- Analysts
-
From each engineer's own setup to one pipeline the team shares.
- Software Engineers
- QA
-
Routine platform work goes to agents, and your team approves every change.
- DevOps
- SRE
- Platform Engineers
-
Specs your engineers' agents can build from.
- Product Managers
- Product Owners
- UX
-
What to steer, what to measure, and what to leave alone.
- CTOs
- VPs
- Engineering Managers
-
Harden your agents, then use them to harden your code.
- AppSec
- Architects
- Software Engineers
Questions before you book
Which coding agents do your trainings run in?
Claude Code, Codex, GitHub Copilot, Gemini CLI and Cursor. We've delivered every engineering, product and leadership training in each of them, and your team trains in the one it already uses.
Which agent should our team choose?
The one your company already licenses and allows. If you don't have one yet, we bring the agent that suits your team, and you pay for the licences. The practice carries over from agent to agent, and in the prep call we tell you where your agent is strong and where it's weaker.
Do you provide a training environment?
Yes: a hosted environment with Claude Code. For Codex, GitHub Copilot, Gemini CLI or Cursor, your team works on its own machines and licences, and your code stays there.
Not sure where to start? Take the self-check