À propos de Pierre-Emmanuel
I work side-by-side with founders to turn bold ideas into production-ready, AI-driven products.
• ML Training to Deployment: End-to-end workflow from model design and training through production rollout.
• Data Engineering: Architecting and maintaining ETL pipelines & databases, plus integrating vector databases for fast, similarity-based search.
• MVP Development: Hands-on building of minimum viable products to validate ideas quickly.
Co-founder @ Recove
Led the complete development (backend & frontend) of Recove—the app that connects Pinterest to secondhand. Recove has 1K+ daily active users and a 50% weekly retention rate, delivering personalized second-hand fashion recommendations straight from users’ fashion inspirations.
Français
Bilingue ou natif
Anglais
Capacité professionnelle complète
Accepte de travailler sur site
Paris (jusqu’à 10 km)
Expériences
- gracqCLTV Prediction & YouTube tutorialE-COMMERCEavril 2025 - avril 2025Introduce a Python library for customer lifetime value (CLTV) prediction and delivering a comprehensive, hands-on tutorial as a YouTube video. My work included:• Library Assessment: Conducted an in-depth review of the library’s installation process, API surface, model implementations, documentation quality, and performance on benchmark datasets.• Tutorial Planning: Outlined a structured walkthrough covering data ingestion, feature engineering, model training, evaluation (e.g., MAPE, RMSE), and deployment considerations.• Hands-On Coding: Developed end-to-end example scripts demonstrating how to load customer data, train and tune the CLTV model, and interpret its outputs.
- RecoveCo-FounderHIGH TECHaoût 2024 - Aujourd'hui (1 an et 10 mois)Paris, FranceRecove is the app that lets users connect their Pinterest account to Vinted and receive personalized second-hand fashion recommendations from their moodboard images. As technical co-founder, I’ve owned both backend and frontend development, and my responsibilities include:• Building a multimodal search engine that recommends second-hand items based on an image• Designing and maintaining the database that aggregates catalogs from multiple brands, and developing automated workflows to keep it continuously updated• Developing the API that powers both the web and native interfaces• Crafting the user interface for seamless, engaging experiences
- HelloSafeLead Data ScienceBANQUE & ASSURANCESjuin 2023 - août 2024 (1 an et 2 mois)Paris, FranceHelloSafe is a web platform offering free insurance product comparison tools. Working closely with the co-founder, my primary mission is to design LLM-based solutions that simplify creation of content on the website.The methodology is divided into two streams:• Database Construction: Build the HelloSafe database from unstructured sources (PDFs, web pages) by combining information-retrieval techniques, image preprocessing, and automatic JSON generation via LLMs.• Automated SEO Content: Leverage LLMs to automatically generate web content aimed at improving our SEO and driving organic traffic.
Recommandations
Soyez le premier à recommander Pierre-Emmanuel
Contribuez à la réussite de ce freelance en partageant votre expérience de collaboration avec lui.
Ces profils de freelance correspondent également à vos critères
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
Formations
- Master spécialisé Data ScienceENSAE ParisEnseignements principaux : - Apprentissage statistique appliquée - Deep Learning - Bases de données - Data storytelling Enseignements optionnels : - Statistique bayésienne - Business data challenge - Cloud computing - Machine Learning avec Python - Machine Learning appliqué à la finance - Apprentissage par renforcement - Blockchain, Bitcoin & Smart-Contracts
- Master Économiste d’entrepriseUniversité de Tours2022Enseignements : - Economie Industrielle, Théorie des Jeux - Machine Learning - Econométrie appliquée - Python, R, VBA, SQL