AI & digital solutions
Don't just talk about AI. Turn it into something useful.
For SMEs, executives and operational managers: custom tools that save time on your real processes, with no magic promises.
Three areas of action
Concrete needs, fitting tools
AI business tools
Put AI to work on your documents, procedures and data.
- Intelligent assistantsE.g. an assistant that writes SQL queries and KPIs for a WMS
- Custom AI agentsE.g. an agent running on your own computer, driven from your phone
- Workflow automationE.g. sorting incoming emails and drafting replies
- Document processingE.g. reading invoices, delivery notes or résumés
- Productivity toolsE.g. Claude, ChatGPT or Copilot for summaries and meeting notes
Web & mobile applications
Applications tailored to your processes, from prototype to daily tool.
- Business applicationsE.g. an in-house CRM used by 110 people
- Internal management toolsE.g. tracking inventory, quotes or schedules
- Interactive dashboardsE.g. logistics and sales KPIs in Power BI
- Customer portalsE.g. online booking and payment
- Prototypes and MVPsE.g. a first version to test within a few weeks
Process digitalization
Leave re-keying and fragile Excel files behind.
- Removing repetitive tasksE.g. no more re-keying between ERP, WMS and Excel
- Replacing manual Excel processesE.g. Power Query, VBA or a web application
- Centralizing dataE.g. one SQL database instead of scattered files
- Automating workflowsE.g. Power Automate, n8n or Make for approvals and reminders
- Intelligent reportingE.g. self-updating reports, alerts on variances
Method
From an idea to a working application
- 01
Understand the business problem
Observe the current process and isolate what really costs time.
- 02
Design the solution
Define the useful scope, the data and the user journey.
- 03
Build the application
Develop in increments, with versions you can try early.
- 04
Test and improve
Check with users and correct based on their feedback.
- 05
Deploy and support
Go live, train people and evolve the tool.
Proof
Applications that really exist
Every project shows its real status: deployed, internal tool, prototype or demonstration.
My approach to AI
- AI assists, and a human validates sensitive actions.
- Confidential data stays under control: a local model is an option when it makes sense.
- No promise of full autonomy: we measure what actually works.
Have an idea? Let's build it together
Describe the process that costs you time. I will tell you honestly what is feasible.