Available for new engagements
David Brunet — CTO consultant CTO consultant, fullstack developer, network & infra architect
For 17 years I have designed, built and run complete platforms: from the first idea to the production launch, and long after.
17+
years of systems in production
180
technicians managed on telecom operations
3
platforms designed and run end to end
48 h
to get a first reply to your message
Services
One person, three hats. You pick the one you need, and the other two come with it.
A product to launch, a team to structure, an architecture to get back on its feet: I bring an outside view, reasoned decisions and a plan your teams can execute.
What I take on
Formats
I build complete applications, from the database schema to the interface, and deliver them with their tests, pipeline and monitoring. No prototype left sleeping in a repository.
What I take on
Formats
I started out running network roll-outs in the field. Today I design and operate cloud-native infrastructure that is reproducible, observable, and stays under your control.
What I take on
Formats
End to end
Most projects get lost in the hand-offs. I hold the whole chain, so nothing falls through the cracks.
Idea
Business scoping
Architecture
Diagrams, choices, costs
Code
Backend, frontend, AI
CI/CD
Tests and delivery
Kubernetes
Deployment
Production
Monitoring, incidents
Work
Three systems I designed, built and still operate.
Sovereign legal AI01
An AI assistant for legal professionals: sourced, verifiable answers, with no dependency on a US model provider.
EdTech ecosystem on Kubernetes02
A learning platform built from open-source building blocks, with a single identity layer and fully owned infrastructure.
E-commerce in production03
A specialised online store, designed and run as a real production system: traffic, orders, customers.
Why me
I managed up to 180 technicians before I made a living from code. I know what a delay, an outage or a vague decision costs, and I explain it without jargon.
I operate what I build. Monitoring, backups, incidents: I design systems I am willing to be woken up by at three in the morning.
No silver bullet, no fashionable technology forced on you. If your problem is first an organisational one, I tell you before billing any code.
Method
Understand the business, the constraints and the budget before talking technology. You leave with a clear scoping note.
A simple, documented architecture where every component is justified. What is not needed does not get in.
Short iterations, a pipeline from day one, and a production release as early as possible.
Monitoring, documentation, training for your teams. The goal: a system that runs without me.
Background
2005 — 2009
Operational management of up to 180 technicians. Field roll-outs, network supervision, customer support and complex incident resolution.
Since 2009
Creating and supporting digital projects: software architecture, fullstack development, infrastructure, security, CI/CD and operations. A long-term technical partner for my clients.
Ongoing
NVIDIA Inception programme, incubation at EuraTechnologies, work with CITC in Lille. Design of RAG systems and self-hosted LLMs.
Today
I run my own platforms in production and support companies on their architecture, infrastructure and AI projects.
Articles
Field notes on architecture, infrastructure and AI.
Rewriting everything is almost always a bad idea. Here are the criteria that led me to do it anyway, and the method for not losing the product along the way.
Reflections on a discussion about Obsidian, second brains, and AI — and the confusion between useful tools, wow effects, and real effectiveness.
A field report on building a sovereign legal AI without dependency on OpenAI, through the LEGIBUS project.
Why most AI projects hit their first obstacles well before the choice of model or technology.
Contact
Write to me. I reply within 48 hours, and the first conversation commits you to nothing.