Claude Code training for your team. Set up, secured, in daily use.
Claude Code is installed quickly. Until a team works with it reliably, it needs shared rules in the repository, clear permissions and verification steps for every change. Then there is the question of what happens to your data. That is exactly what we set up with you in the workshop – on your own code.
- 3–12 people
- 1–2 days
- In-house
- With certificate
01 — Who it is for
For teams that want to use Claude Code seriously. Not for a product demo.
Teams introducing Claude Code
The licences are ordered or already distributed. Now this should become a shared way of working, before twelve private variants take hold.
Tech leads and CTOs
You have to define what the agent may do, which servers it can reach and how its changes are verified – and defend that to data protection and the works council.
Platform and DevOps teams
Claude Code should not only run on the laptop but also in the pipeline: as a reviewer, for recurring tasks or headless in CI.
02 — What your team learns
Introducing Claude Code in your team: eleven topics in four groups.
We decide in the intro call which topics form the focus. Each topic ends with an exercise whose result stays in the repository.
01 — Basics
CLAUDE.md and AGENTS.md
Project memory at team, project and folder level. Shared rules in AGENTS.md, included via import so that other tools read them too.
Context and token management
What the model currently sees, when compacting or a fresh context pays off, and how to estimate the cost per task.
Plan mode
Plan first and review the plan, then let the agent make changes. The simplest lever against agents that confidently build the wrong thing.
02 — Extend
Skills
Describe recurring workflows as a skill: creating a migration, preparing a release, writing a test case.
Subagents
Separate contexts for research, tests and review, with restricted tools and a clear assignment.
Hooks
Deterministic rules before and after every tool call: formatters, tests, blocked paths. What must always apply does not belong in a prompt.
03 — Connect
MCP servers
Connect ticket system, documentation and database – configured project-wide, with vetted servers and minimal permissions.
Headless and CI
Use Claude Code without an interface in scripts and pipelines, for example for reviews of pull requests.
04 — Secure
Permissions and sandboxing
Define allowed and forbidden commands in the project, isolate shell access, distribute company-wide policies centrally.
Review and test rules
Which checks every change passes before it may enter the main branch – and where a human approves.
Data protection and hosting
Anthropic directly or via Amazon Bedrock and Google Vertex AI, EU regions, contract terms: what applies to your data.
03 — What the repository contains afterwards
No slide deck. A configuration that stays.
At the end of the workshop, your repository is set up for Claude Code. The files are versioned, agreed within the team and apply to every run – on the laptop as well as in the pipeline.
- AGENTS.md
- rules, architecture and commands for all agents
- CLAUDE.md
- includes AGENTS.md, adds Claude-specific settings
- .claude/settings.json
- permissions and hooks, versioned in the repository
- .claude/skills/
- recurring workflows of the team
- .claude/agents/
- subagents for review, tests, research
- .mcp.json
- approved MCP servers for the project
04 — Process and key facts
From intro call to certificate. In one to two days.
One possible structure for two days: on the first day basics, CLAUDE.md, plan mode, skills and subagents. On the second day hooks, MCP servers, headless and CI, as well as permissions, review rules and data protection.
01
Intro call
We talk about your stack, tools, prior experience and what currently causes friction in daily work. Then we select the building blocks and clarify laptops, licences and approvals with your IT.
02
Workshop
One or two days on site at your company. Short theory blocks, followed by exercises on the sample CRM or on your own code. From six participants, two trainers are in the room.
03
Wrap-up
All participants receive a certificate. What was created in the workshop – rules, skills, verification steps – stays in your repository and keeps working the next working day.
Optional
Follow-up support and consulting
On request, we support the first weeks afterwards: we look at how rules and verification steps work in daily practice and refine them. If you want to change your development process as a whole, you will find that in our AI consulting.
- Location
- In-house
- On site at your company
- Group size
- 3–12 people
- From 6 people with two trainers
- Duration
- 1–2 days
- Depending on topics, recommended: 2 days
- Preparation
- Briefing call
- Topics matched to your team
- Format
- Hands-on
- Theory, applied in practical exercises
- Completion
- Certificate
- For all participants
05 — How the two differ
Claude Code training or AI training? A question of the tool.
Claude Code training
One tool in depth, for teams that have settled on Claude Code.
- Configuration, skills, subagents and hooks in detail
- Permissions, sandboxing and central policies
- Use headless and in CI
AI training for developers
The way of working, independent of the tool – also for GitHub Copilot, Cursor and Codex.
- Spec-driven development and good practices
- RAG, MCP and ticket workflows
- For teams with several tools in use
06 — Why Viewconic
We work with Claude Code ourselves every day. The workshop comes from this work.
At Viewconic, we build business software for clients with coding agents. Claude Code is one of our main tools. What we show in the workshop comes from exactly this work: the same rules and verification steps – and the experience they came from.
What this process looks like in projects, we describe under AI-assisted software development. If you want to change your development process as a whole, our AI consulting can help.
“David from Viewconic stands out for outstanding quality of results and an agile way of thinking that lets us deliver even demanding client projects efficiently and with focus. The team works professionally, is solution-oriented and always brings the right mix of expertise and flexibility.”
References
- Carl Zeiss
- SAP
- Munich Re
- atra.consulting
Companies our team has developed for, partly before the founding of Viewconic. The training series is run together with atra.consulting.
07 — Frequently asked questions
Questions about the Claude Code training. Licences, data, duration.
What does a Claude Code training cost?
After the intro call, you receive a fixed-price quote for the entire workshop. It depends on duration, group size and whether one or two trainers are in the room.
Does every participant need their own Claude licence?
Yes. Each person needs access to Claude Code, for example through a Team or Enterprise plan from Anthropic, through the API, or through Amazon Bedrock or Google Vertex AI. Which route fits your requirements, we clarify in the intro call, and we coordinate the setup with your IT in advance.
Can we practise on our own code, and what happens to the data?
Yes, that is the usual way. Alternatively, we work on a sample CRM. We use your company’s own accounts, so your existing contract terms apply, and we clarify in advance which repositories are approved. Production data is not required.
Is one day enough?
One day is enough for the basics, CLAUDE.md, plan mode, and skills and subagents. If hooks, MCP servers, use in CI and the security rules are to be added, we recommend two days.
We also use GitHub Copilot or Cursor. Is a Claude Code-only training worthwhile then?
If several tools are in use, our vendor-neutral AI training for developers usually fits better. The basics transfer anyway: rules in AGENTS.md, specification first and quality gates work with any tool. The Claude Code training goes into depth on one tool.
Request Claude Code training. Or book an intro call directly.
Tell us how many people will take part and how far Claude Code is already in use at your company. Or book an intro call directly.