About Caroline
Senior Data Scientist (10+ years) ML & GenAI specialist. Built and deployed end-to-end ML solutions on AWS.
Strong in advanced Python, OpenAI/CrewAI/GoogleADK, time-series forecasting, BigQuery, SQL, and. experience working in cross-functional teams to deliver measurable business impact in operations and marketing.
Open to Lyon-based role (onsite 2-3 days/week).
English
Fluent
French
Native or bilingual
Can work on-site
Lyon (up to 10km)
Experience
- ConfidentielData scientistE-COMMERCEMarch 2022 - May 2025 (3 years and 2 months)Lyon, France
- Built production forecasting pipeline (Prophet + external regressors) achieving 15% MAPE. Automated training and deployment with CI/CD , integrated output into AdWords budget allocation workflows to improve campaign ROI. Tech: Python (pandas, NumPy), Docker, Git, CI/CD, BigQuery → (deployed to AWS SageMaker or equivalent),
- Designed geospatial analysis using H3 + Python to evaluate retail network coverage; developed a Streamlit decision-support app for field teams, enabling targeted expansion and reducing operating costs (estimated €500K). Tech: Python, H3, geopandas, Google Place API, Streamlit, BigQuery.
- Developed GenAI pipeline (LLaMA / ChatGPT) to extract and classify GDPR sanctions from CNIL reports; packaged as a production Streamlit app for compliance teams with model versioning and basic monitoring. Included data preprocessing, evaluation, and human-in-the-loop validation. Tech: LLaMA, OpenAI APIs, Python, AWS SageMaker, Streamlit, Docker..
- Designed and delivered ETL workflows and ingestion for competitor monitoring Tech: Scrapy Playwright, AWS Sagemaker jobs, BigQuery, Streamlit, pytesseract), reducing data latency to near real-time
- ITG ConsultantsB2B Data Analyst (free-lance)MECHANICAL ENGINEERINGJanuary 2018 - February 2022 (4 years and 1 month)Paris, France
- Led data quality, enrichment, and optimization initiatives for account targeting based on technographic and firmographic criteria → Delivered ICP modeling for clients such as Wedia, BirdOffice, and PG Organisation in collaboration with Eleius Marketing
- Conducted digital performance benchmarking across global markets for clients including Michelin Group (TyreDating) and Prevision.io → Identified optimization levers through comparative analysis and positioning diagnostics
- Designed and implemented client segmentation models to improve targeting, messaging, and commercial alignment → Supported strategic pivots for companies like Qonto and Disruptive Innovation Group, via classification models and actionable scoring matrices
- GL eventsData Marketing ManagerApril 2005 - September 2017 (12 years and 5 months)Lyon, FranceStrategic Project Support: Provided data-driven support for large-scale tenders and international events: World Expo, FIFA World Cup, etc. •
- Led data collection, market intelligence, and competitive analysis to guide strategic positioning and proposal writing
- Developed custom scraping workflows to extract and structure high-volume unstructured data (press, web, tender databases)
Marketing Insight Automation & Scalable Tools : •- Automated the generation of marketing insights for event benchmarking, enabling effective segmentation, targeting, and opportunity scoring for commercial teams
- Designed and implemented a search engine scraping over 10 million web pages per year, adexing event data by industry sector, region, and performance metrics (tool used to respond to +500 web-based data requests annually, significantly improving commercial responsiveness)
Acted as a key interface between business stakeholders and external data providers, ensuring strategic relevance and data quality across all initiatives.
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
- Concepteur développeur en sciences de donnéesJedha20221-Construction et alimentation d'une infrastructure de gestion de données (scraping, AWS, cartographie) 2-Analyse exploratoire, descriptive et inférentielle de données (data visualisation) 3-Analyse prédictive de données structurées par l'intelligence artificielle : machine learning supervisé (prédiction / classification / clusterisation par Kmeans et DBScan) 4-Analyse prédictive de données non-structurées par l'intelligence artificielle (NLP : Modèles de deep learning séquentiels et transformers) 5-Industrialisation d'un algorithme d'apprentissage automatique et automatisation des processus de décision 6-Direction de projets de gestion de données : rétroengineering des facteurs de ranking sur Google
- The Hacking ProjectThe Hacking Project2017Bootcamp de développement sur Ruby on Rails : HTML, CSS, Javascript
Certifications
- Data SEO LAb Niveau 1Data SEO LAB