APR 13, 2026
At the Intersection of Engineering and Applied Intelligence.
Developing robust methodologies and production-grade systems for high-stakes AI, with a focus on methodology, clinical NLP, high-quality synthetic data generation, and LLM-driven systems.
Selected Works
Engineering Projects
CLINICAL NLP EVALUATION
2026 / PUBLISHED
PARHAF CliBench
A benchmark for measuring how 7B-9B language models handle structured information extraction on real French clinical notes.
View Case Study east
REALTIME AUDIO TRANSCRIPTION
2026 / ONGOING
Asifonix
A low-latency ASR system thought for multiuser real-time applications in controlled environments.
LICENSE PLATE RECOGNITION
2022 / PUBLISHED
GetThePlate
End-to-end automatic license plate recognition system based on deep learning.
Primary Focus Areas
Systematic inquiry into the underlying mechanics of intelligent systems.
01
Reliable AI Systems
Building controlled LLM, RAG and agentic systems with validation evaluation, and deterministic safeguards.
02
Clinical Decision Support
Designing AI and NLP systems that support clinicians, streamline workflows, and improve patient care.
03
Custom AI Systems
Developing tailored systems for domain-specific needs in modeling, probabilistic reasong, and decision workflows.
04
Applied Research
Creating rigorous evaluation setups and open tools that help others study, test, and extend systems.
The Journal
Recent Field Notes
NOV 12, 2025
MAR 13, 2024
Professional
Trajectory
Bridging the gap between theoretical computer science and industrial engineering.
Data Scientist / Applied ML Engineer
2024 - PRESENTHealth Data Hub
Full-stack ownership across a live healthcare data platform, from ingestion pipelines to deployed predictive models. Contributed to several internal tooling initiatives alongside the core modelling work, including documentation automation and structured data generation.
Data Scientist
2022 - 2023SogetiLabs (part of Capgemini)
Worked across several applied ML problems: clinical NLP extraction, EEG-based biomarker classification, and LLM-assisted test automation. A broad mandate, handled with a consistent focus on business impact and interpretability.