EZÀ propos de Edwin
Curious and adaptable, I’ve built and deployed models across a from computer vision for drones and plant health detection, to time series forecasting for energy systems, to signal processing for security. I love solving real-world problems, especially those that have strong impact : health, sustainability, or innovation.
🔹 Strong in Machine Learning, Deep Learning, and Generative AI
🔹 Hands-on with Computer Vision (YOLOv7, Mediapipe, image classification/segmentation)
🔹 Proficient in Python, PyTorch, TensorFlow, scikit-learn
🔹 Experience with Signal Processing (QPSK), SARIMA
🔹 Built full data pipelines, dashboards, and Streamlit apps; connected to APIs (SharePoint)
🔹 Comfortable with SQL, Pandas, Geopandas, Git, Linux
I enjoy learning on the fly, getting up to speed with new tools or data, and building something that actually gets used. I’m also keen on working within an interdisciplinary team and love to learn from others.
Let’s build something that matters.
Français
Bilingue ou natif
Polonais
Bilingue ou natif
Anglais
Capacité professionnelle complète
Japonais
Notions
Espagnol
Notions
Accepte de travailler sur site
Marseille (jusqu’à 50 km)
Expériences
- NepTech,Data Scientist Internavril 2025 - Aujourd'hui (1 an et 2 mois)Marseille, FranceBuilt relational database with automated SharePoint API synchronization for real-time data ingestion from CFD results. Developed energy consumption prediction models for electric boat fleet using ensemble architectures, achieving 8% MAPE for ship resistance calculations and reduced worst-case scenarios. Deployed Streamlit application enabling non-technical staff to generate energy forecasts, reducing CFD simulation time for pre-project calculations. Implemented model validation framework and conversational AI interface for dataset exploration.
- Doshisha University,Data Scientist Internseptembre 2023 - février 2024 (5 mois)Kyoto, JapanWorked on QPSK modulation and computed encoding/decoding error in message transmission based on a paper research, including a study of numerical stability and estimation of key parameters.
- 3L-Optronics,Data Scientist Internmars 2023 - septembre 2023 (6 mois)Marseille, FranceDeveloped a semi-supervised learning method for the recognition of diseased apple tree leaves, combining YOLOv7 for detection and classification algorithms like Efficient Teacher on 3000 images, with a comparison of two approaches to optimize performance.
Recommandations
Soyez le premier à recommander Edwin
Contribuez à la réussite de ce freelance en partageant votre expérience de collaboration avec lui.
Ces profils de freelance correspondent également à vos critères
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Formations
- Master degree in EngineeringCentrale Marseille Engineering School2025Master degree in Engineering
- Information and Data ScienceDoshisha University2024Information and Data Science
Certifications
- Transformers Hugging Face CourseHugging Face2025
Compétences
Catégories
- Autre