My name is Rémy Giraud. I design and deploy artificial intelligence solutions for industrial SMEs: a company memory that answers from your documents, citing its sources, then agents that take on the tasks that cost you the most. Custom-built solutions, under your control, judged on one thing only: results measured in your company.

I did not choose this job on a whim. It comes from 8 years spent in industry, with one common thread: finding where a company loses value, and getting it back.

What industry taught me

For 8 years, I have been working in the aerospace and defense industry, where I worked my way up from apprentice to Operational Excellence Manager. Behind the title, a very concrete job: tracking down what slows a plant, what costs too much, what produces errors, then fixing it, with measurements to back it up. I practiced it across 2 very different sites. First a large, highly structured production plant in France, with the standards you would expect from a large group. Then a 250-person SME in Switzerland, a subsidiary of the same group, where my scope stretched far beyond production: sales, finance, R&D, every department in the company. That is where I discovered from the inside how an SME runs, and understood that it is not a scaled-down version of a large group.

An engineer by training, certified Lean Six Sigma Black Belt, I learned two things there that never left me: never claim anything without a measurement, and when a team lacks a digital tool, build it myself rather than wait for it to appear.

A large part of that job was also about passing knowledge on: I trained and coached entire teams, from the shop floor to the offices. It taught me that an improvement only sticks if the people who live with it make it their own. That is even more true with AI, and it is why training your teams is part of every deployment.

And everywhere, the same observation: what holds a company back is almost never the competence of its teams, it is knowledge that is out of reach at the very moment it would pay off. The engineer who could win the tender spends his days hunting down references instead of writing the quote. The answer that would reassure a customer waits on one expert, who is pulled in every direction. The analysis already paid for once gets redone, because nobody can find it. Behind these ordinary scenes, there are deals won or lost, margins wearing thin, talent wasted on searching for information. That is when I understood where the biggest source of untapped value in a company lies: not in yet another tool, but in the knowledge it already owns, put back to work.

The turning point

When artificial intelligence started showing up everywhere, what I mostly heard was noise: immense promises, very little proof. I wanted to understand. How these tools really work, and where they can bring value. I taught myself: hundreds of hours of research, reading and experimentation, to separate what works from what only makes for nice demos.

Out of that work came a clear vision, built on 2 levels: a company memory as the foundation, then automation with measurable impact on the processes that cost the most. And one conviction: deploying AI is not about installing software, it is about improving a process. The projects that succeed start from a precise problem, measure before, prove after. It is the job I have been doing for 8 years. I simply changed tools.

My way of working

• Measure before promising. The diagnostic quantifies your current situation before any commitment: you decide on facts, not on my enthusiasm.
• Start small, prove, extend. A first scope, a review based on numbers, then we expand.
• Transparency as a rule. Direct conversations rather than grand presentations, and a frank opinion on your situation, even when it does not go my way.