You're seeing this page as if you were . The main menu is still yours, though. Exit from immersion
Sébastien BouilléSB

Sébastien Bouillé

Reliable AI Systems Architect | FR · EN · ES

480 €/día
Barcelona, ES
3-7 años

Tiempo medio de respuesta: 1h

Acerca de Sébastien

You want to use AI — but you're afraid of hallucinations,
of ruining your system or your credibility.

You're right to be.

I build AI systems that don't drift. The reliability is
in the architecture, not in the prompt: the AI works within
constraints I design, and every output is validated before
it reaches the user.

What I implement:
- Constrained inputs — RAG, JSON or Markdown rule systems
- Constrained outputs — validation scripts that check every
answer before delivery
- Fallback logic and guardrails for edge cases
- Self-maintaining systems that flag inconsistencies when
your code or content changes

If you need to keep the AI's creativity, I help you add the
right guardrails to get as close as possible to zero
hallucination — with human-in-the-loop where it matters.

I've built three AI systems from the ground up:
- Plumeum — a LinkedIn ghostwriting app with self-improving
rules
- Lunema — an adaptive meditation app, zero hallucinations
in production
- Mon Histoire Magique — multilingual generative stories
for children

Before AI: ten years as an international journalist (trained
to make complex things clear and defensible) and seven years
in technical support for a major US tech company.

If you want AI that works reliably — not just in the demo,
but in production — let's talk.
  • Francés

    Bilingüe o nativo

  • Inglés

    Competencia profesional completa

  • Español

    Competencia profesional completa

Acepta trabajo presencial
Barcelona (hasta 30 km)

Experiencia

  • Independant
    Reliable AI Systems Architect | FR · EN · ES
    AGENCIAS DE SUBCONTRATACIÓN
    agosto de 2024 - Hoy (2 años)
    Barcelona, España
    AI systems that use the right data and deliver constrained, reliable outputs. systems I design are self-maintaining and self-updating: part of the code or content changes, they flag the inconsistency instead of failing silently. AI's implement guardrails hallucination — with human-in-the-loop
  • US tech company (confidential)
    Technical Support Advisor (T2)
    AGENCIAS DE SUBCONTRATACIÓN
    enero de 2017 - enero de 2024 (7 años)
    Barcelona, España
    Smart home, connected hardware, and voice assistant products (client confidentiality). Frontline diagnosis at international scale; contributor to documentation and faster-resolution workflows.
    Training
  • Independant
    International field journalist
    CINE & AUDIOVISUAL
    enero de 2008 - enero de 2017 (9 años)
    Doha, Catar
    Ten years of TV field reporting for French and international broadcasters (M6, TF1 , France 2, Al Jazeera, France 24, Arte, RTS). Trained to translate complex situations into clear, defensible output under pressure. Author of two published novels.

Recomendaciones

Sé el primero en recomendar a Sébastien

Ayuda a este freelance a destacar compartiendo tu experiencia.

Estos perfiles de freelance también coinciden con tus criterios

AgathaA

Agatha Frydrych

Backend Java Software Engineer

4.7

(3)

2

BaptisteB

Baptiste Duhen

Fullstack developer

4.6

(4)

5

AmedA

Amed Hamou

Senior Lead Developer

4

(2)

7

AudreyA

Audrey Champion

Web developer

4.3

(3)

4

Formación

  • DUT Services et réseaux de Communication
    Université de Dijon
    2004
    Programming, Network, Computer graphics

Certificados

  • Supervised Machine Learning — DeepLearning.AI
    Stanford University
    2025
  • Elements of AI
    University of Helsinki
    2024

Conjunto de habilidades profesionales

Categorías