Freelancer profile translated to English.

Description

🎓 Young graduate from Télécom SudParis, I am a junior developer specializing in Data Science and Python, passionate about solving real-world problems through data.

💡 My goal: gain experience by participating in real projects, while seriously contributing my technical expertise.

🚀 Skills:

Data analysis, cleaning, visualization (Pandas, Matplotlib, Seaborn)

Machine Learning/Deep learning (scikit-learn, tensorflow)

SQL, Python

Tools: Jupyter, Git, Streamlit, etc.

📁 I have already worked on several academic and personal projects (see GitHub) and I am curious, rigorous, and a fast learner.

✅ I am available for short or long-term assignments, and my daily rate is adaptable to your needs.

Industry field of expertise

Languages

English
Native or bilingual
French
Native or bilingual
Spanish
Conversational

Workplace preferences

Can work on-site

Paris (up to 50km)

ALTEN
AI engineering
RESEARCH
July 2025 - Today (1 year and 1 month)
Sèvres, France
R&D on generative AI applied to computer vision
Pedestrian detection and tracking for autonomous vehicle road safety
Implementation of generative approaches based on VLM fine-tuning and multimodal RAG
GPU memory optimization (offloading, LoRA fine-tuning, NumPy preprocessing optimization for CPU-GPU transfers)
Implementation of non-generative approaches, creation of a CNN-LSTM pipeline for pedestrian detection and tracking
Model benchmarking in dynamic contexts (urban environment, partial pedestrian occlusions)
Computer Vision & Image Processing Machine Learning LLM Retrieval-Augmented Generation (RAG) VLM
Kaggle
Data analyst
E-COMMERCE
January 2025 - March 2025 (2 months)
Paris, France
Exploratory analysis of e-commerce customer data to identify behavioral patterns and business levers
Complete e-commerce data analysis pipeline: ingestion, cleaning, EDA, and visualization
Design of KPI dashboards for customer behavior analysis and decision support
Customer segmentation based on RFM metrics and unsupervised clustering
Temporal study of purchasing behaviors (recency, loyalty, seasonality)
Data Cleaning and Preprocessing Data Science Data Analysis BI Strategy Pandas
Kaggle
Data scientist
BANKING AND INSURANCE
March 2025 - May 2025 (2 months)
Paris, France
Design of a supervised machine learning pipeline for fraud detection
Training and comparison of several models: Random Forest, XGBoost / LightGBM
Model selection based on Recall, Precision, F1-score, and AUC-PR rather than accuracy
Handling severe class imbalance using resampling techniques (oversampling, SMOTE, and class weighting)
Hyperparameter optimization to improve fraud detection while limiting false positives
Machine Learning Data Science EDA Data Cleaning and Preprocessing