Freelancer profile translated to English.
Back to original languageAbout Mohamed
I meet your Data & Artificial Intelligence needs thanks to a dual expertise in AI and Data, acquired through my two specialized master's degrees and my professional experiences:
• Understanding and scoping: needs analysis, definition of AI/Data use cases, strategic scoping.
• Data exploitation: exploratory analysis, statistical modeling, machine learning & deep learning, scoring and forecasting methods (classification, regression, time series).
• Advanced AI: NLP (LLM, RAG, fine-tuning), Computer Vision (classification, detection, generation), recommendation and personalization.
• Technical implementation: software development, management and optimization of massive databases, industrialization and deployment in production, model performance monitoring.
French
Native or bilingual
English
Fluent
Can work on-site
Lyon (up to 50km)
Experience
- Institut Français du Pétrole Energies nouvelles (IFPEN)AI EngineerENERGY AND UTILITIESSeptember 2025 - Today (11 months)Lyon, France• Development of AI tools based on LLMs, multi-agent systems, and RAG.• Automation of documentary transposition and writing of technical documentation.• Multi-agent systems for business process automation.• Development of an intelligent quoting assistant based on specifications.Stack: Python, LangChain, LangGraph, RAG, OCR, Dash, Azure AI Services
- Kheopsys (Client : Appodeal)Data Scientist ConsultantCONSULTING AND AUDITSMarch 2025 - August 2025 (5 months)Paris, France• Design of advertising recommendation systems to optimize app installations.• Identification of similar users through clustering and lookalike embeddings.• Prediction of click-through rate (CTR) using Deep Learning and Machine Learning models.• Data processing pipelines with AWS, SageMaker, and Athena.Stack: Python, Scikit-learn, PyTorch, FuxiCTR, AWS, SageMaker, Athena, SQL, Polars
- STMicroelectronicsAI EngineerNANO-TECHNOLOGYMarch 2024 - August 2024 (5 months)Crolles, France• Intelligent classifier to evaluate the relevance of design software errors (PDK rules).• Optimization of LLM encoder models by hyperparameter tuning and fine-tuning.• Complete pipeline: preprocessing, training, inference, retraining, and evaluation.• Recall ~100% on test sets, validated by a microelectronics engineer.Stack: Python, Scikit-learn, PyTorch, TensorFlow, Hugging Face, NLTK, Spacy
Recommendations
Be the first to recommend Mohamed
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
- Data & Connected SystemsUniversity Jean Monnet & School of Mines of Saint-Etienne2026Master 2
- Machine Learning & Data MiningUniversity Jean Monnet2024Master 2