Tell us about your project
Claude training in Luxembourg is delivered by AIxH as a module of its AI training programme, approved as a continuing vocational training organisation by the Luxembourg Ministry of the Economy (authorisation 10156897/1). The module is detailed below; formats, prices and co-funding for the whole programme are grouped on the AI training & upskilling page.
Claude, developed by Anthropic, is less familiar to the general public than ChatGPT. Its strengths lie elsewhere: reading a long document in full, drafting and rewriting over time while keeping a consistent style. Those uses matter first to the professions that handle large volumes of text: law, finance, insurance, human resources, communications. All of them are well represented in Luxembourg. A second, more recent side has been added: Claude is the tool that goes furthest in putting AI to work inside your files and your software, through its agents and connectors. The training covers both.
Programme
The fundamentals shared by every module (LLM basics, prompt engineering, ethics and governance, confidentiality and GDPR included) are described on the programme page. For Claude, the uses below come on top, practised on the documents and cases participants bring along: reading and writing first, then the agents that execute.
Claude has the lowest hallucination rate on the market and a built-in caution: when it does not know, it says so. Its nuanced writing makes it the favourite tool of text-heavy professions. On a short question, a generalist will do; on a case file or a report, the difference shows.
Have Claude read a contract, a report or a complete case file, and draw out a summary, a list of points to watch or the answer to a specific question.
Write a first draft, recast it in a given register, harmonise the style of a document written by several hands. Custom styles record the company's register and reapply it from one document to the next. Exercises are run on letters, minutes and internal reports.
Alongside the conversation, Claude generates artifacts: a formatted document, a comparison table, a web page, a small interactive tool. The result is shared through a simple link, with nothing to install. The workshop produces some on participants' own cases.
A Claude project gathers the reference documents, specific instructions and conversation history of a case file or a job function. Uploaded once, the documents serve every question that follows. The workshop builds one for the team and configures the tool's memory.
Claude connects to Drive, your inbox, your calendar or in-house software through the MCP protocol, created by Anthropic and since adopted by the rest of the market. The workshop shows what a connector changes day to day and how to frame the access rights.
Cowork gives Claude a workstation: drop in a folder of receipt photos and it produces an Excel expense report with formulas and a chart. An agent built for recurring, scheduled assignments. To place the tool against its OpenAI competitor, read our analysis of ChatGPT Work vs Claude Cowork. On what Claude really costs in agent mode, see our analysis of Claude Fable 5.1 pricing.
The Chrome extension lets Claude act directly on websites: filling in a form or extracting the data from a page while you do something else. In Excel, it turns a raw file into a structured table. Both are demonstrated live during the workshop.
Describing a need is enough to get a first working internal tool: a calculator or a dashboard, for instance. Claude Code builds the application, Claude Design polishes its layout. The workshop shows the principle on a simple case, without writing a line of code.
Who it is for
This module is aimed first at teams whose work rests on reading and writing documents: law firms, funds and management companies, insurers, HR, communications. More broadly, any team that regularly produces long written material or has to summarise it. No technical background is needed to follow the workshop.
Formats and prices
Like the other modules in the programme, the Claude training runs as a half day, a full day, an in-house session or a conference talk. Prices and co-funding are brought together on the AI training & upskilling page.
Full programme
Four other tool modules complete the programme:
FAQ
Confidentiality and retention guarantees vary with the type of account (free, Pro, Team, enterprise). The session starts from the configuration the company actually uses to say what can go through it and what must stay outside, as an extension of the programme's GDPR strand.
That depends mostly on the volume of documents to process in the workshop: a free account is enough to discover the tool, but working on contracts or long reports generally calls for a paid subscription, and so do the agents (Cowork, Claude Code). The exact tier is chosen with the team before the session.
The way the workshop runs does not change from one module to the next (real cases, approval, co-funding). The content does: writing reliability, long documents and agents that work in your files here, personalising the assistant (memory, custom GPTs, projects) on the ChatGPT training.
Yes, it is one of the tool's strong points: it reads an entire document in one go, without splitting it into chunks. The workshop draws directly on participants' own case files.
Cowork is Claude's agent that works directly in your folders and files: it can turn photos of receipts into an Excel expense report, prepare a deliverable or run a recurring assignment end to end. The workshop demonstrates it and identifies the team's tasks that lend themselves to it.
No. It is the tool that goes furthest in building a prototype from a plain description: an internal calculator or dashboard, for example. The training shows the principle on a simple case; taking it into production remains a separate integration job.
Yes, through connectors based on the MCP protocol created by Anthropic: Drive, email, calendar or in-house software. The session shows connectors in use and the access framework to define before opening up the data.
Yes. The Claude module is covered by the same co-funding schemes as the rest of the programme, detailed on the AI training & upskilling page.
Client reviews
★★★★★ 5.0 / 5 on Google (43 reviews) · See our Google profile →
Reviews were left on Google in French and are translated here.
“AIxH delivered a high-quality training course on artificial intelligence. Grégory and Kenza are approachable trainers who listen and share, with a tailored approach. We learned a great deal. Thank you!”
“Excellent training and intelligent support that leads to real skills. Thank you.”