Victor Sonck
R&D Data Scientist AI For Recycling
Ghent Metropolitan Area Belgium
Summary
👤
Senior🎓
Top SchoolVictor Sonck is an R&D data scientist and developer advocate with a decade of experience applying ML and MLOps to real-world problems, currently focused on AI for recycling at VITO. He blends a business-engineering background with deep technical hands-on skills (Linux, Python, Web, R) and has contributed to prominent open-source projects like ClearML and YOLOv5, improving experiment tracking, data versioning and CI/CD workflows. As a ClearML Developer Advocate and former ML6 engineer, he moves fluidly between building integrations, improving developer UX, and speaking publicly about ML solutions across Europe. Based in Ghent, Victor also brings board-level perspective through nonprofit work and a penchant for tinkering—often surfacing practical conveniences (like local HPO and faster dataset versioning) that make ML teams more productive.
10 years of coding experience
8 years of employment as a software developer
Master's degree Data Analytics, Master's degree Data Analytics at Ghent UniversityCollege Degree Economics, College Degree Economics at Sint-Lodewijkscollege Lokeren
Dutch, English, French
Stackoverflow
31reputation
2kreached
2answers
1question
Github Skills (34)
editors10
pytorch10
docker10
qt10
python10
spyder10
machine-learning10
editor10
dockers10
cicd10
automation10
deeplearning-ai10
pyqt10
deep-learning10
mlops10
Programming languages (10)
Github contributions (5)
clearml/clearml
Dec 2021 - Dec 2022
ClearML - Auto-Magical CI/CD to streamline your AI workload. Experiment Management, Data Management, Pipeline, Orchestration, Scheduling & Serving in one MLOps/LLMOps solution
Role in this project:
MLOps Engineer
MLOps EngineerContributions:2 reviews, 6 commits, 8 PRs in 11 months
Contributions summary:Victor contributed to the ClearML repository by adding features to support machine learning workflows and improving the user experience. They added options for local hyperparameter optimization, enabling users to run optimization tasks directly on their machines. The user also added convenience functionality to the `clearml-data` module, allowing for faster dataset version creation, and integrated git credentials for the colab example. In addition, they provided an example of a CI/CD integration using ClearML, demonstrating the user's understanding of deploying and automating ML tasks.
streamlineawsdata-managementversion-controldata-science
ultralytics/yolov5
Aug 2022 - Dec 2022
YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
Role in this project:MLOps Engineer
MLOps EngineerContributions:3 reviews, 6 commits, 6 PRs in 4 months
Contributions summary:Victor primarily contributed to integrating and improving ClearML experiment tracking within the YOLOv5 framework. Their work involved implementing ClearML's data version management, hyperparameter optimization, and image logging capabilities. They also addressed compatibility issues, enhanced documentation, and refactored ClearML integration for improved performance and maintainability. Further contributions included ensuring that the "best.pt" model file is preserved, mapping project and task names, and adding docker info for remote execution.
pytorchdeep-learningyolov8object-detectionmachine-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.