Summary
Paul Boniol is a researcher at Inria working at the intersection of massive data series management, anomaly detection, and machine learning for time series, with a decade of research and engineering experience across top French institutions. He completed a PhD in collaboration with EDF R&D on detecting anomalies and their precursors in large time series collections, and has held postdoctoral and research-engineer roles at École Normale Supérieure, Université de Paris, and École Polytechnique. His work blends practical industrial applications (e.g., EDF) with foundational methods such as unsupervised subsequence anomaly detection, streaming detection, and discriminative subsequence explanation. Based in Paris, he pairs strong mathematical training from Grenoble INP ENSIMAG and Trinity College Dublin with hands-on systems experience, making him comfortable moving between algorithm design and production-ready analytics. An often overlooked trait is his sustained focus on precursor identification—bridging detection with interpretable causes—which makes his research particularly valuable for real-world monitoring and diagnosis.
10 years of coding experience
3 years of employment as a software developer
Master 2 (M2), Information systems engineering, Master 2 (M2), Information systems engineering at Ecole Nationale Supérieure d'Informatique et de Mathématiques Appliquées de Grenoble
Doctor of Philosophy - PhD, Mathematics and Computer Science, Doctor of Philosophy - PhD, Doctor of Philosophy - PhD, Mathematics and Computer Science, Doctor of Philosophy - PhD at Université Denis Diderot (Paris VII)
Mathématiques et informatique, Mathématiques et informatique at Trinity College Dublin
English, French