Arnaud is a data-driven problem solver with a strong analytical mindset and a knack for turning complex challenges into actionable insights. With experience in both technical and strategic roles, with a focus on Natural Language Processing and Large Language Models (LLM’s), he bridges the gap between data science and business impact. Passionate about continuous learning and collaboration, he thrives in dynamic environments where he can drive innovation and efficiency.
💼 Professional experience
Agilytic
Data Scientist Since 2021
- Health insurance Collaborated in a cross-functional team, including platform engineers, data engineers, RPA developers, and data translators, to create innovative client-focused solutions using AI. All projects were realized using cloud-based technology and version control through Github.
- Developed an AWS solution using various NLP algorithms (LLM, text similarity, embeddings, named entity recognitions, …) and computer vision algorithms (object detection, image classification, …) to extract information from unstructured documents (PDFs, emails, images, etc.) and automate their processing
- Developed algorithms to classify documents based on their visual and textual features
- Developed an NLP classification algorithm to match clients' emails with template responses
- Fine-tuned deep learning models to classify images
- Used AWS cdk to push and deploy containerized applications
- Actively used the AWS cloud environment including EC2, S3, Lambda, SNS, SQS, Aurora, ECR, Cloudwatch, Sagemaker, etc.
- Retail - RAG and AI Agents Developed assets, documentation and testing framework to support the deployment of RAG and AI Agents apps
- NGO Extraction and analysis of data from OpenStreetMap to evaluate cycling infrastructure accross EU countries and regions. Gathering of key metrics on cycling paths and infrastructure quality, creating visualizations and geographical plots to assist with strategic decision-making.
- Retail Development of two churn models for different client segments, with profit lift modeling to optimize retention actions
- Professional Services Development of 3 cross-sell models for different product lines, with direct subsequent actions being tested.
- NGO Development of Power BI dashboards to monitor KPIs of programs.
Technologies: Python, SQL, AWS services, AWS cdk, LLM, HuggingFace Transformers, Pytorch, Tesseract, Docker, Git, spacy, CamemBert, OpenCV, Linux
Technologies: Python, CrewAI, LangGraph, OpenAI, Claude, Dify, Docker
Technologies: Pyhton, Overpass, OpenStreetMap, Geopandas
Technologies: Python for data processing and modeling (pandas, sklearn, xgboost)
Technologies: Python for data processing and modeling (pandas, sklearn, xgboost, SMOTE), SQL
Technologies: Power BI, Excel
B12 Consulting
Data Science internship 2020
Developed a tool retrieving information from scanned invoices through optical character recognition technologies (OCR). He then trained an anomaly detection model to identify in real time suspicious data detected by the OCR.
Technologies: Python (pandas, sklearn, pyod, tensorflow, pytesseract) , Docker, Bitbucket, FastAPI.
dFakto
Senior Business Consultant 2016-2019
Analyzed business needs on data management projects and Business Intelligence projects for Belgian and international clients in various sectors. He identified and translated business needs to propose budgeted solutions and planned them according to agile principles. Arnaud also built reports and dashboards using visualisation and BI tools.
Technologies: Qlik (Qlik Sense, QlikView), JIRA, Odoo
🦾 Certifications, trainings
Methodologies
- Machine learning and AI
- LLM applications development
- RAG, Agentic RAG, Agents
- Natural Language Processing
- Optical character recognition (OCR)
- Computer Vision
- Data Engineering principles
- Dashboarding & Visualisation
- Agile developments
- Popularization of technical topics for non-technical audience
- Communication and presentation to senior management
Software & programming
- Python, R
- LLM, HuggingFace, OpenAI API, CreawAI, langchain, Langgraph
- Pytorch, Tensorflow, sklearn, transformers
- Tesseract, AWS Textract
- spacy, CamemBERT, NLTK
- OpenCV, scikit-image
- AWS
- Docker
- SQL
- Git
- Linux OS
Certifications
- AWS Solutions Architect Associate
- Azure fundamentals
- Microsoft Power BI
- Prince2 Foundation
- SAS base programming
🎓 Academic credentials
- Msc in Data Sciences (Statistics) UCLouvain, 2021 Master’s thesis : Anomaly detection using copula-trees
- MSc in Business Engineering UCLouvain, 2016
🇺🇳 Languages
🇫🇷 French: native
🇬🇧 English: C1
🇳🇱 Dutch: B1
🔐 Proprietary and confidential. 💡 Profile CVs are provided for illustration purposes. ✅ Planning of specific experts will be secured upon formal agreement.