Freelancer profile translated to English.
Back to original languageAbout Ouijdane
Data Scientist with 2 years of experience, I design solutions based on data and artificial intelligence to help companies analyze their data and detect patterns useful for decision-making.
I have notably worked on data analysis and fraud detection projects within the Groupe La Poste, combining data analysis, machine learning, and log analysis.
My missions included:
- Data analysis and creation of interactive dashboards with Microsoft Power BI to support decision-making.
- Exploration and analysis of logs via Splunk to identify fraud and identity theft scenarios.
- Development of machine learning (MLP) models for the detection of abnormal behaviors.
- Vectorization of user interactions and behavioral analysis with NLP techniques (Word2Vec, Sentence Transformers).
- Implementation of statistical analyses to assign anomaly scores and detect behavioral deviations.
- Experimentation with solutions based on LLMs and generative AI for intelligent log analysis.
I am particularly interested in projects related to data science, data analysis, machine learning, and AI applied to anomaly detection and business process optimization.
French
Native or bilingual
English
Fluent
Can work on-site
Paris (up to 50km)
Experience
- Pôle Data IA - La Poste GroupeData ScientistPUBLIC SECTOROctober 2025 - Today (10 months)Fraud Detection:Development and implementation of a solution based on LLM (Large Language Models) technologies using generative AI to analyze logs and detect suspicious behavior.
- Pôle Data IA - La Poste GroupeData Scientist (Apprenticeship)PUBLIC SECTORSeptember 2023 - September 2025 (2 years)Nantes, FranceData Analysis with Power BI:
- Development of a project for HR on the analysis of a survey on the impact of a move.
- Creation of interactive dashboards with Power BI.
- Presentation and adaptation of dashboards for end-users.
Fraud Detection:- Exploration, analysis, and collection of logs and identification of identity theft scenarios using Splunk and based on the Mitre Attack framework.
- Development of an AI model based on a multilayer perceptron (MLP) to classify users based on their daily activities and detect suspicious users.
- Model training on a large machine via GitLab, implementation of a continuous integration (CI/CD) workflow.
- Analysis of user searches and manipulated files to vectorize suspicious behaviors detected by NLP techniques such as Word2Vec, Sentence Transformer.
- Development of a vectorization model to convert user interactions into numerical representations exploited by fraud detection techniques.
- Implementation of advanced statistical methods to analyze variations in usage behaviors and assign an anomaly score, allowing the identification of deviations from expected behaviors.
Technical Environment: Power BI, Splunk, Python, PyTorch, GitLab CI/CD, VS Code
Recommendations
Be the first to recommend Ouijdane
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
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
Education
- Engineering degree in data and artificial intelligenceESEO Vélizy-Villacoublay2025Diplôme d'ingénieur en data et intelligence artificielle
- DEUST in mathematics, computer science, physicsFaculty of Science and Technology of Fez Fez2022
Certifications
- Microsoft Azure AI Essentials Professional Certificate by Microsoft and LinkedInMicrosoft2026