London, UK • EMEA
French
English
AI/ML Engineer building production enterprise agents, governed RAG systems, and ML pipelines across cloud environments. I work from stakeholder discovery and architecture through integration, evaluation, deployment, security, observability, adoption, and technical enablement.
Available immediately for London-based opportunities and open to roles across EMEA. Get in touch
Available immediately for London-based opportunities and open to roles across EMEA. Get in touch
Junior World Champion| Prague, 2019
2x Vice-Champion of France| 2024, 2025
Top 15 World Ranking| Current
Work Experience
Decathlon FranceSept 2025 - Present
AI Engineer- Led client discovery meetings and end-to-end delivery of DAISI (Decathlon AI Supplier Informations), a production Google Chat Enterprise Agent for supplier and finance workflows covering Workday Finance integration, internal APIs, invoice management, purchasing procedures, and internal organizations. Designed for a 10,000-user target population with a reported annualized impact of 13,000 hours/year saved.
- Implemented DAISI as a ReAct-style agent with LangChain/LangGraph, FastAPI, Gemini and Vertex AI, Vector Database and Search, Cloud Run, Cloud SQL, MLflow/Databricks, Model Armor, DLP, GDPR safeguards, and Terraform. Optimized cost, latency, scale, security, caching, and runtime reliability while preserving auditable, source-backed behavior.
- Validated production readiness through integration tests, detailed technical reports, security controls, evaluation traces, and a successful load test with 2,000 concurrent users.
- Designed A2A-ready interoperability interfaces and built an internal Agent Factory with reusable libraries, templates, tool interfaces, traces, evaluation, observability, security, and deployment standards, improving agent delivery time-to-market by up to 5×.
- Led OpsBot, an Enterprise Agent for maintenance, safety, and compliance workflows. Built governed-source RAG with source-backed answers, safe fallback, MLflow traces, evaluation, Cloud Run, Vertex AI RAG Engine, Terraform, GDPR safeguards, and the shared agent library. Expected impact and ROI hypothesis: 14,500+ h/year reallocated and ~€540k productivity reallocation.
- Led Summarizer, a Gemini-powered summarization platform orchestrated with Airflow/MWAA, AWS EKS/S3/ECR, Databricks Delta refresh, runtime validation, and operator runbooks.
- Presented production agent-development practices to 100+ colleagues and delivered 30+ hours of GenAI, AI, and Tech training covering architecture, evaluation, security, observability, and cloud delivery.
- Led internal AI/Tech innovation through a weekly newsletter, onboarding guide, documentation/runbooks, and internal hackathons.
LangChain
LangGraph
RAG
Gemini
Vertex AI
GCP
Cloud Run
Cloud SQL
Terraform
FastAPI
MLflow
Databricks
Airflow/MWAA
AWS EKS
S3
ECR
Docker
PostgreSQL
Decathlon BelgiumMay 2025 - Aug 2025
Machine Learning Engineer- Led Belgium Forecast, a strategic machine learning project that translated stakeholder forecasting needs into production predictions for 8 key sales KPIs (including GMV and items sold), segmented by channel (InStore/OutStore, 1P/3P) across 64 sports categories.
- Built an automated forecasting pipeline spanning Prophet and XGBoost, Apache Spark, PySpark, Databricks, Delta Lake, AWS S3, Airflow/MWAA, and MLflow, replacing a manual process with reproducible training and delivery.
- Achieved a 15% forecast improvement, measured with MAPE, versus the previous manual process through rigorous comparison of Prophet, XGBoost, LightGBM, and Chronos-Bolt, plus weather, holiday, and lag feature engineering and parallel processing with joblib.
- Deployed models to MLflow Model Registry for production inference, automated Google Sheets and Tableau exports for business stakeholders, and built GitHub Actions CI/CD, SonarCloud quality gates, and Sphinx documentation.
Prophet
Apache Spark
Databricks
MLflow
AWS S3
Airflow
GitHub Actions
Python
Beobank NV/SAMay 2023 - Aug 2023
Data Scientist- Processed and analyzed banking transaction data with Python, VerticaPy, NumPy, Pandas, SQL, JupyterLab, and Excel, investigating classification failures and translating recurring patterns into new transaction categories and business rules.
- Improved transaction categorization success from 63% to 84% through iterative quality analysis, while supporting income and expense identification and reusable reporting for digital banking analytics.
Python
SQL
Pandas
NumPy
VerticaPy
JupyterLab
Excel
PowerPoint
Education
JUNIA ISEN - Lille, FranceEngineering Degree, AI & Big Data (2023 - 2026)
Ranked 2nd/190 in final year - Top 1%. Specialization in AI, Big Data, Data Science & Machine Learning.
Data Structures & Algorithms, Machine Learning, Deep Learning, Distributed Systems, Database Management (SQL & NoSQL), Operations Research, Computer Networks, Big Data (Hadoop, Spark), Cloud Computing, DevOps, Computer Vision, NLP, Software Architecture
Preparatory Classes (2021 - 2023)
Ranked 2nd/96 (Year 1) • 8th/76 (Year 2) · Highest Honors
Ranked 2nd/190 in final year - Top 1%. Specialization in AI, Big Data, Data Science & Machine Learning.
Data Structures & Algorithms, Machine Learning, Deep Learning, Distributed Systems, Database Management (SQL & NoSQL), Operations Research, Computer Networks, Big Data (Hadoop, Spark), Cloud Computing, DevOps, Computer Vision, NLP, Software Architecture
Preparatory Classes (2021 - 2023)
Ranked 2nd/96 (Year 1) • 8th/76 (Year 2) · Highest Honors
Jean Perrin High School - LambersartHigh School Diploma (Baccalauréat) (2018 - 2021)
With Highest Honors
Mathematics, Physics-Chemistry, Computer Science
With Highest Honors
Mathematics, Physics-Chemistry, Computer Science
Technical Skills
AI EngineeringPython, LangChain, LangGraph, RAG, Vector Databases (Vertex AI, Qdrant, Pinecone, Weaviate), evaluation, observability, load testing, caching, safety, guardrails, data protection, privacy, memory, Agent2Agent (A2A), and Gemini Enterprise Agent Platform.
Python
LangChain
LangGraph
RAG
Vector Databases (Vertex AI, Qdrant, Pinecone, Weaviate)
Evaluation
Observability
Load Testing
Caching
Safety & Guardrails
Data Protection & Privacy
Memory
Agent2Agent (A2A)
Gemini Enterprise Agent Platform
Machine LearningForecasting, Fine-tuning, PyTorch, TensorFlow, scikit-learn, XGBoost, LightGBM, MLOps, MLflow, and Weights & Biases.
Forecasting
Fine-tuning
PyTorch
TensorFlow
scikit-learn
XGBoost
LightGBM
MLOps
MLflow
Weights & Biases
Cloud & DevOpsGCP, AWS, Azure, Docker, Terraform, CI/CD, FastAPI, SonarCloud, and Wiz.
GCP
AWS
Azure
Docker
Terraform
CI/CD
FastAPI
SonarCloud
Wiz
Big DataApache Spark, PySpark, Databricks, SQL, PostgreSQL, NoSQL, Firestore, MongoDB, and Redis.
Apache Spark / PySpark
Databricks
SQL
PostgreSQL
NoSQL
Firestore
MongoDB
Redis
TeamworkGit, GitHub, Jira, Confluence, Atlassian, and Agile/Scrum.
Git
GitHub
Jira
Confluence
Atlassian
Agile/Scrum