DIGITAL TECHNOLOGY IN PRACTICAL TERMS
Ecommerce Manager, a developer at heart.
I manage e-commerce projects on a day-to-day basis, and I code the tools that go with them. Here, I document the things that really interest me: cybersecurity, automation, AI and data.

4 angles of attack
e-commerce, development, IA, automatisation
My method
My Approach
Start with the numbers, not gut feelings
Nine years in e-commerce have taught me one thing: convictions are worthless without data to back them up. Before any decision — redesigning a purchase funnel, reallocating budget, launching in a new market — I start by measuring. GA4, Power BI, P&L tracking: data is the starting point of every project, and the final judge at the finish line.
Automate everything that can be automated
Time spent on repetitive tasks is time stolen from strategic thinking. Reporting, price monitoring, KPI tracking: I automate systematically, whether with off-the-shelf tools or workflows I build myself (n8n, APIs, scripts). It's as much a professional conviction as a personal passion — I spend part of my free time building automations and experimenting with new tools.
Understand the tech to communicate better
Trained in front-end development alongside my e-commerce career, I speak the language of developers as well as that of marketing. This dual culture prevents misunderstandings, speeds up projects, and makes it possible to challenge technical solutions rather than simply accept them.
Test, measure, iterate
I'd rather run ten small, measured experiments than make one big blind bet. CRO, acquisition, pricing: every hypothesis gets tested, every result analyzed, every lesson fed back into the loop. That's how you make lasting progress — across 11 markets or just one.
Stay curious, always
AI, cybersecurity, new data tools: digital moves fast, and I consider continuous learning an integral part of the job. What starts as personal exploration today often becomes an operational advantage tomorrow.
Areas of expertise
E-commerce
Développement Front-end
Cybersécurité
Automatisation & IA
Featured analysis
Cybersécurité
MAI-Cyber-1-Flash: Microsoft Bets on AI to Industrialize Cybersecurity
Microsoft has launched MAI-Cyber-1-Flash, its first cybersecurity-focused AI model, integrated into MDASH. By pairing this compact model (handling 90% of tasks) with heavier models for edge cases, Microsoft claims a 96% score on the CyberGym benchmark while cutting costs by 50%. The announcement also introduces Perception, an agentic system for continuous threat monitoring. A concrete illustration of smart model-routing based on task complexity.
E-commerce
AI Overviews in France: What It Really Means for Advertisers
AI Overviews and AI Mode rolled out in France in late July 2026, but ads embedded directly inside these AI blocks aren't live there yet — only classic Search and Shopping ads appear around them. CPCs are rising and clicks are getting scarcer, pushing advertisers to target higher up the funnel instead of focusing only on bottom-funnel, transactional keywords. Automation (PMax, AI Max) should be adopted gradually, not as a full replacement for traditional Search — going all-in too fast risks losing control over performance. Tracking and data quality (a clean CRM, Consent Mode V2, server-side tracking) are becoming prerequisites for Smart Bidding to work properly. Ad performance increasingly depends on overall marketing strategy: brand awareness, content, and multichannel presence.