Three days. A more productive way to use AI.

Why this programme, why now

Your dossiers are moving faster than any inbox

The regulatory files keep moving, the monitoring and the content never stop, and the team that produces both is small.

• An audit first. We start by mapping how the team uses AI and software today. Then we review the tools and LLMs available, how to connect them with the tools you already use to raise productivity, and how to link the team's accounts so everyone works on the same setup.
• Day one: setup and direction. The first day focuses on your environment and your connections. We look for the pain points worth improving, then identify the skills and tools that fit your case.
• Compliance built in. On day one we also go through what Article 4 of the AI Act asks for, and we integrate it into the shared memory we build during the programme.

The programme at a glance

Days 1 and 2 run back to back. Day 3 can follow immediately or sit one to two weeks later, so the team arrives with real questions.

Day 1 · The audit sprint

We look at the real work first

No slides and no hypothetical cases. We open the editorial calendar, the inbox and the drive, and build the programme around what we actually find.

What we cover

• How you work today. We map how the team uses its current tools, review the settings, and define the best case scenario to aim for.
• Claude, set up properly. We configure Claude for the team and pick the skills that fit your case.
• NotebookLM and the LLM landscape. A practical look at Google NotebookLM and at which model to reach for per task.
• The shared memory. We set up an Obsidian vault as the team's shared memory: terminology, house tone, stakeholders to watch, topics to flag.
• Connecting everything. Claude connected to that memory and to Google NotebookLM, so every tool works from the same context and the team's accounts stay aligned.
• The AI Act, together. We go through what Article 4 asks for and integrate it into the shared memory.
• The council. We build the council of AI agents that reviews the team's output before it goes out.

Day 2 · Creative production

The content engine, hands-on

A full day at the keyboard for the two people who produce FEVE's content. Everything is built on real documents and saved as reusable templates, so the methods survive the workshop.

What we cover

• Claude, connected to your tools. We connect Claude to the programs the team already uses, so it works inside your real workflow.
• Claude Design. How to set it up and use it to produce visuals, layouts and campaign assets in house style.
• Your avatar. We create an AI avatar for the team and show how to use it for video content.
• Voice with ElevenLabs. We set up ElevenLabs and add voice to your content.
• Social media automation. We build the connections to automate posts on X and LinkedIn through their APIs, and walk through the whole process live.

Day 3 · Base automation

One job stops being manual, in front of everyone

The wider team joins for the day the method becomes a machine. We take the worst-scoring job from day 1 and build the automation live, with your data and your tools. Not a prototype: the real thing, still running after we leave.

What we cover

• The automation that matters most. We pick together the automation with the biggest impact on the team's week, based on what came out of day 1.
• The automation software, installed. Introduction and setup of the automation software, and the technical part: connecting all the APIs it needs.
• Design together, build after. We draw the structure of the automation together in the room. The technical parts that take longer are built by us afterwards, on that design.
• Routines. A second type of automation: recurring routines that run on their own, on a schedule.
• A new flow from scratch. A third type: a fresh flow built end to end, for example a scraping flow that turns daily coverage into a press review.

Before we start

A few questions before day one

Tick what is already true for your team. Your answers tell us where to start, and nothing here is required in advance: whatever is missing, we set up together.

0 / 7 checked

The AI Act, handled

The training is also the evidence

Article 4 of the AI Act asks organisations to support the AI literacy of their staff, and to be able to show what measures they took. This programme produces that evidence as a by-product of the work: an attendance-based register with names and dates, documented workflows linking real AI use to human review, and a written operational policy.

One programme, two results: a faster team, and a file you can put on the table if anyone ever asks.

Investment

Clear numbers, one invoice

Prepared by White Rabbit Brussels for FEVE. Sessions run on site in Brussels, in English; materials can be delivered in additional languages on request. This proposal is valid for 30 days. Full programme details: sprints.brusselswhiterabbit.eu