I work as a data scientist in the digital production group and I am part of the DataDrive team. I am involved in several project and I am managing the computer vision coordination project (a SFF11 subprojet).
Data science contains a broad spectrum of activities and it is a challenging field, not only because it comes at the interface of computer science and math, but also because:
There are a lot of available tools in the market.
Working in data science involves having some domain knowledge (often different than our domain knowledge of comfort)
The reproducibility and generalizability of created systems and models
My main function is to mitigate these difficulties in a research context by contributing to data and machine learning aspects in the projects I am contributing to. Here are few examples of the projects I am working on:
Computer vision coordination project: enable and foster computer vision based projects at agroscope by supporting CV based projects and organizing regular meetings and workshops about the topic.
Rumex detection using computer vision: in this project, I work on the modelling and the system architecture.
Smart irrigation using dendrometers signals over LoraWAN: in this project, I work on signal and data analysis and on implementing the backend routine for automation.
Rhythmicity as welfare indicators for ruminants: I develop algorithms (in the context of the DigiRhythm package) and analyze data.
Google Scholar: https://scholar.google.fr/citations?user=L97ZODwAAAAJ&hl=en