About Raphael
Your data is worth more than you're currently getting out of it!
🎯 AREAS OF EXPERTISE
🛠️ TECHNICAL STACK
- Clustering and classification
- Topic modeling
- Sentiment analysis
- Time series forecasting
- Recommendation engines
- Anomaly detection
- Python, Scala, PySpark, SQL, R
- Data manipulation: Pandas, NumPy
- Machine Learning: Scikit-Learn, XGBoost, LightGBM, MLLib
- Deep Learning: Keras, PyTorch, TensorFlow
- Graphs: NetworkX
- Vector databases: PostgreSQL (pgvector), Pinecone, Haystack
- Frameworks: LangChain, LlamaIndex
- LLM APIs: Claude (Anthropic), OpenAI, Cohere, Mistral, OVHcloud AI
- Google Cloud Platform: BigQuery, Vertex AI, Cloud Functions, Cloud Run
- Azure: Azure ML, Azure Functions
- Relational: MySQL, PostgreSQL, SQLite, Hive
- NoSQL: MongoDB
- Graphs: Neo4j, Gremlin
- Plotly, Matplotlib, Seaborn, Tableau, Apache Superset
- Requests, Selenium, BeautifulSoup, Scrapy
- Front-end: React, HTML, CSS
- Back-end / APIs: FastAPI, Flask
- Reflex, Streamlit, Dash
- Git, Docker, CI/CD
- MLflow, model monitoring
French
Native or bilingual
English
Native or bilingual
Spanish
Fluent
Experience
- TotalEnergiesLead Data ScientistENERGY AND UTILITIESJanuary 2023 - Today (3 years and 7 months)Paris, FranceSales Forecasting:Developed and deployed a prediction model (N-HiTS) on GCP, improving forecast accuracy by ~50%. Automated pipeline and visualization interface for business teams.SEO Semantic Clustering:Created an algorithm grouping +700k Google impressions by search intent, enabling SEO teams to optimize their content strategy. Deployed via Streamlit on GCP.Customer Comment Monitoring:Implemented a topic modeling tool (BERTopic) analyzing +10k comments to identify trends. Automated classification of new comments by semantic proximity, with LLM labeling (Gemini). Interactive dashboard for monitoring.Digital Asset Generation:Developed a web application for creating visuals integrating AI image generation (Gemini, Imagen) and a visual editor. Reduced production time by several days compared to agencies.
- Johnson & JohnsonCustomer Engagement Analytics ManagerPHARMACEUTICALS INDUSTRYJanuary 2022 - December 2022 (1 year)Paris, FranceManaging the adoption of an omnichannel recommendation platformInterfacebetween business and technical teams to align business needs and product development. Accompanied 6 teams in deployment. Produced analytics reports identifying engagement drivers and increasing platform adoption.
- Hewlett Packard Enterprise (HPE)Data EngineerSOFTWARE PUBLISHINGSeptember 2021 - January 2022 (4 months)Paris, FranceMigration and industrialization of data pipelines:Migrated pipelines from R to PySpark on Dataiku. Corrected legacy code flaws, strengthening infrastructure reliability.
Recommendations
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
- Bachelor of Science (BSc) in Information Systems and ManagementUniversity College London2016
- Master of Science (MSc) in Data ScienceKing's College London2018
Certifications
- Neural Networks and Deep LearningCoursera2019
- Mentor on OpenClassroomsOpenClassrooms2018