
TINGFEN YU
Data Scientist & Engineer
Salaire / Taux journalier
78500 Sartrouville, France
Freelance, ouvert au CDI
Expertises
Langues
À propos
Introduction
Data Scientist & Engineer with 5+ years building ML models and scalable data pipelines at companies like Deezer and Pikkopay. Specialized in clustering algorithms, NLP, and behavioral analytics using Python, Spark, and GCP. Expert in turning complex datasets into actionable business insights that drive 15% improvements in user satisfaction.
Expérience de travail
Data Scientist
Pikkopay
avr. 2025 - Présent
1 an 7 mois
Paris, France
• Clustering Model Development: Designed and implemented clustering models to segment users and detect behavioral similarities, with the goal of optimizing customer experience and personalization strategies. • Behavioral Data Analysis: Analyzed user behavior data collected via Sentry to enrich machine learning models and identify key usage patterns, enabling targeted improvements to product features and user engagement. • Interactive Dashboards and Monitoring: Built dynamic dashboards to visualize key user metrics such as retention, acquisition, and activity, providing actionable insights for product and marketing teams. • Machine Learning and NLP Techniques: Explored supervised learning models and natural language processing techniques to deepen behavioral insights, and contributed to the design of scalable data pipelines for predictive analytics.
Data Scientist / Engineer
Deezer
sept. 2023 - mars 2024
7 mois
Paris, France
In my role as a Data Scientist Intern at Deezer, I'm involved in projects aimed at enhancing discography navigation through the refinement of canonical albums. My primary missions include: • Data Consolidation and Deployment: Conduct data consolidation of internal and external sources using Scala Spark, implement unit tests, and deploy it on the Google Cloud Platform (GCP). • Data Pipeline Development: Create end-to-end data pipelines for the ingestion, processing, and storage of large-scale datasets, using Spark/Scala, PySpark, BigQuery, Python, GitHub, and Docker to ensure efficient and scalable data workflows • Machine Learning Approach: - Apply a combination of TF-IDF algorithm and linear-assignment algorithm to get the ground truth. - Explore various machine learning approaches, including agglomerative clustering, random forest, and LGBM, to identify the optimal model for canonical albums. - Implement random forest model and GrahX to group the similar albums and refine canonical albums, followed by production deployment by creating jobs in PySpark. - Optimize algorithm using NLP's locality-sensitive hashing (LSH) method for blocking to speed up predictions in production.
Data Analyst
Tarena International Inc
févr. 2015 - juil. 2017
2 ans 6 mois
Shanghai, Chine
As a Data Analyst at Tarena International Inc, I played a pivotal role in investigating complex business problems using data and analytics tools such as dataiku, esensoft etc. My responsibilities include: Customer Journey Analysis: Conducted in-depth customer journey analysis using RFM and AIPL framework. Leveraged insights to optimize customer engagement strategies, enhance satisfaction, and drive retention. I have contributed to a 15% improvement in user satisfaction by refining the personalization of user paths. KPI Monitoring: Established and implemented robust dashboards to monitor Key Performance Indicators (KPIs). This involved creating visual representations of data to facilitate real-time decision-making and enhance overall business performance. Strategic Development: Collaborated with cross-functional teams in the formulation and execution of strategies for products and services. Contributed to the development of comprehensive plans to ensure alignment with organizational goals.
Quality Control Supervisor
ZHUBO DESIGN
févr. 2014 - janv. 2015
1 an
Shanghai, Chine
I was in charge of collaborative platform. My main mission was to orchestrated seamless collaboration on architectural projects by fostering an environment conducive to efficient communication and streamlined workflows among team members.
Electrical Engineering Specialist
华东建筑设计研究总院 ECADI
févr. 2011 - janv. 2014
3 ans
Shanghai, Chine
Key Responsibilities and Achievements: - Building Information Modeling (BIM) Expertise: Spearheaded the implementation of BIM methodologies within the electrical engineering domain, streamlining collaboration between architectural and engineering teams. Leveraged cutting-edge BIM software to create accurate digital representations of electrical systems, enabling comprehensive analysis and optimization. - Electrical Infrastructure Design: Led the development of intricate schematics for electrical infrastructure, ensuring compliance with local regulations and industry standards. Collaborated closely with architects, structural engineers, and other stakeholders to integrate electrical systems seamlessly into overall building designs, optimizing space utilization and functionality. - Quality Assurance and Compliance: Implemented rigorous quality assurance measures to guarantee the reliability and safety of electrical systems. Conducted thorough inspections and audits to identify potential issues and ensure compliance with relevant designs and standards, mitigating risks and ensuring the longevity of installed infrastructure.
Formation
EPITA: Ecole d'Ingénieurs en Informatique
Engineer's degree
2021 - 2024
4 ans
Tongji University
Bachelor's degree
2007 - 2011
5 ans