◆ AI & Innovation
I design and deploy AI solutions within the corporate Microsoft environment. Audit, Copilot, business assistants, automation. No empty promises: tools I master, in a framework I know.
01 AI as a production accelerator
I integrate generative AI tools into my day-to-day development and engineering work. Legacy code analysis, PowerShell / Python / SQL script generation and industrialisation, technical documentation, rapid prototyping, diagnosis of complex incidents: AI lets me deliver faster and explore more broadly, without cutting corners on rigour.
Human control stays systematic. Architecture, technical choices, code review and final validation are my expertise. AI proposes, I decide.
Architecture and technical decisions remain mine. AI accelerates, it does not replace the critical reading of a senior developer.
No sensitive client data is sent to public models without an appropriate contractual framework. Azure OpenAI or local solutions for sensitive contexts.
Reduced timelines, more comprehensive deliverables, controlled costs. The client measures the result, not the method.
02 AI services
I identify the AI use cases that make sense for your business, and those that won't. Opportunity mapping, gain estimation, prioritisation by return on investment.
Deliverable Prioritised, costed AI roadmap
Deployment plan tailored to your organisation, licence management, team training, data governance (DLP, sensitivity labels). Copilot works from day one, and securely.
Deliverable Copilot deployed, governed, adopted
Design and development of AI assistants for your teams — Copilot Studio, Azure OpenAI, custom .NET solutions. Connected to your data, integrated into Teams or your intranet.
Deliverable Functional assistant in production
Power Automate combined with AI Builder to automate business processes that depend on text or image understanding. Mail processing, document classification, data extraction.
Deliverable Automated, measured workflows
03 AI integration in the Microsoft ecosystem
My specificity: AI within the corporate Microsoft environment, not AI in general. I work with the official building blocks, their governance constraints and the security practices that go with them. What sets me apart from generalist AI providers is precisely this mastery of the Microsoft framework.
04 Frequently asked questions
No, never without an appropriate contractual framework. For sensitive contexts, I use Azure OpenAI Service hosted in Europe: your prompts and data don't leave the European border and aren't reused to train other models. GDPR is respected natively, and depending on the case, HDS certification or SecNumCloud sovereignty requirements can be met through the right subscriptions and hosts.
No. AI augments existing teams' productivity; it doesn't replace them. My services are designed as assistants: your people keep the decision-making, AI saves them time on production. The ROI comes from shifting human value to higher-expertise tasks — less mechanical work, more analysis and steering.
Three main lines: Microsoft licences (Copilot, Power Platform), Azure consumption (OpenAI models billed by usage), and integration. I provide a transparent estimate from the framing stage, with low and high ranges depending on expected adoption. Most business assistants cost less than teams think — the real challenge is properly sizing Copilot licences to avoid over-deployment.
4 to 8 weeks depending on the size of the organisation and the maturity of the existing Microsoft 365 governance. The technical deployment takes a few days; the rest is data preparation (sensitivity labels, DLP), training and adoption support. Without those steps, Copilot works but doesn't deliver on its promises: teams don't know what to ask it, or worse, expose data they shouldn't have.
◆ Let's work together