09/14/2026 updated
SL
100 % available
AI/ML Engineer & Data Scientist | Applied Research
Liege, Belgium
Only remote
Master degree in Computer Science, University of LiegeAbout me
AI/ML Engineer and Data Scientist with experience spanning applied research and industrial machine-learning projects since 2020. Strong Python background in modelling complex real-world data, time-series analysis, feature engineering, experimental design and model validation.
Artificial IntelligenceData AnalysisComputer VisionData IntelligencePython (Programming Language)Machine LearningNatural Language ProcessingFeature EngineeringData SciencePyTorchDeep LearningScikit LearnUnsupervised Learning
Machine Learning & AI
Expertise in supervised and unsupervised learning, deep learning, reinforcement learning, and GANs. Proficiency with frameworks such as PyTorch, JAX, TensorFlow, and scikit-learn, including model evaluation and validation.
Data Science & Statistics
Strong background in exploratory data analysis, feature engineering, time-series analysis, clustering, dimensionality reduction (PCA, t-SNE), correlation and factor analysis, model validation, and experimental design.
Programming & Data
Proficiency in Python, SQL, R, Java, C#, MATLAB, C/C++, and Julia, applied to building and deploying machine-learning pipelines and data-driven solutions.
Applied ML Domains
Hands-on experience in time-series modelling, computer vision, NLP, and signal/sensor data analysis, including feasibility studies and PoC development.
Engineering & Research Tools
Practical use of Git, Linux, SLURM, and HPC workflows to support large-scale experiments and reproducible research.
Computer Vision
Design and evaluation of computer-vision pipelines for real-time data extraction, image segmentation, and object recognition under challenging conditions such as overlap, occlusion, and depth variation.
Natural Language Processing
Development of NLP solutions including Named Entity Recognition models, thesaurus construction, and custom word embeddings using Word2Vec and GloVe.
Reinforcement Learning
Design of simulation environments and ML experiments to investigate emergent behaviour of reinforcement-learning agents, with large-scale experimentation on HPC infrastructure.
Expertise in supervised and unsupervised learning, deep learning, reinforcement learning, and GANs. Proficiency with frameworks such as PyTorch, JAX, TensorFlow, and scikit-learn, including model evaluation and validation.
Data Science & Statistics
Strong background in exploratory data analysis, feature engineering, time-series analysis, clustering, dimensionality reduction (PCA, t-SNE), correlation and factor analysis, model validation, and experimental design.
Programming & Data
Proficiency in Python, SQL, R, Java, C#, MATLAB, C/C++, and Julia, applied to building and deploying machine-learning pipelines and data-driven solutions.
Applied ML Domains
Hands-on experience in time-series modelling, computer vision, NLP, and signal/sensor data analysis, including feasibility studies and PoC development.
Engineering & Research Tools
Practical use of Git, Linux, SLURM, and HPC workflows to support large-scale experiments and reproducible research.
Computer Vision
Design and evaluation of computer-vision pipelines for real-time data extraction, image segmentation, and object recognition under challenging conditions such as overlap, occlusion, and depth variation.
Natural Language Processing
Development of NLP solutions including Named Entity Recognition models, thesaurus construction, and custom word embeddings using Word2Vec and GloVe.
Reinforcement Learning
Design of simulation environments and ML experiments to investigate emergent behaviour of reinforcement-learning agents, with large-scale experimentation on HPC infrastructure.
Languages
EnglishFluentFrenchGoodDutchBasic knowledgeRussianFluentSpanishNative speaker
Project history
Development and evaluation of machine-learning models for predicting biomechanical outcomes from multivariate time-series data. Exploratory analysis using PCA and t-SNE, comparison of neural-network architectures with LOSO validation, and collaboration with biomechanics researchers to ensure scientifically meaningful results.
Led an independent feasibility study for automated inventory counting in an industrial warehouse environment. Designed and evaluated multiple computer-vision PoCs under challenging conditions and recommended an alternative weight-based solution after experimental results showed limitations of image-based approaches.
Designed simulation environments and ML experiments to investigate emergent behaviour of reinforcement-learning agents. Implemented experimental pipelines in Python on HPC infrastructure using SLURM, compared model behaviours across conditions, and contributed to interpretation, reproducibility, and methodological validation.