SmartWatch
Scraping pipeline for Lyon public facility opening hours. Leverages embeddings and LLM to produce structured, unambiguous output, compared against data.grandlyon.com.
Python LLM Web Scraping Embeddings NLP
Data Scientist
& AI Engineer
I design and deploy AI-powered solutions and data analysis pipelines with over 20 years of experience, from software engineering to data science. Expert in NLP, LLMs and data pipelines, I combine scientific rigour with a pedagogical approach to turn business challenges into robust, production-ready applications.
Projects
StellaScript
Local Python audio transcription pipeline with speaker diarization, usable in real time (microphone) or on file. Works offline after initial model download.
Python Speech Processing WhisperX Diarization Pyannote SpeechBrain Open Source
PRISM
Composable Python library for string similarity matching. Supports edit distance, sequence similarity, token-based, phonetic and semantic similarity with a unified API.
Python NLP Embeddings String Matching Library
ShadowLog
AI-powered log analysis platform using clustering and statistical anomaly detection to identify patterns in large-scale log files.
Python Machine Learning Clustering Anomaly Detection ELK Stack
ASR.lab
Benchmarking platform for automatic speech recognition systems: controlled audio degradation, enhancement, normalization and multi-engine comparison with interactive reports.
Python ASR Benchmark Whisper Wav2Vec2 Speech Recognition Open Source
ForzaEmbed
Python benchmarking framework for text embedding models: grid search over chunking strategies and similarity metrics, with textual heatmap and embedding space visualizations.
Python NLP Embeddings Benchmark RAG Chunking Open Source
Selma
TypeScript template for visualizing and interactively exploring hierarchical data structures.
TypeScript React Vite D3 Tailwind i18n
School of Statistics
Interactive visualizations for exploring statistical and machine learning concepts.
JavaScript Statistics Data Visualization D3