ARaymond: SmartGlove
Application Engineer, Data Scientist · ARaymond, Grenoble · 2021–2023
Context
An internal startup incubator at ARaymond, set up to translate scientific AI projects into products that could be deployed in the field.
The problem
Machine health engineering: detecting and characterising anomalous behaviour on production lines from multi-source sensor data: audio and inertial sensors mounted on production line equipment.
What I built
A staged pipeline combining anomaly detection, unsupervised clustering, and supervised classification. Separately, I worked on optimising Design of Experiments (DoE) via reinforcement learning, to reduce the number of physical tests needed to qualify a product.
The outcome I care most about was a shift from blind data acquisition to a qualified, controlled acquisition strategy, which improved detection performance and reduced DoE cost. Getting there meant working across sensor installation, modelling, and stakeholder communication. The work I most enjoyed was on the embedded sensors on the line itself.
Stack
Python · MongoDB · Azure · Docker · Git/GitHub