À propos de Matyas
Français
Bilingue ou natif
Anglais
Capacité professionnelle complète
Russe
Notions
Expériences
- MeeticData AnalystRÉSEAUX SOCIAUXoctobre 2017 - septembre 2018 (11 mois)Paris, France- Data extraction, processing and visualization for business analysis (conversion rate, customer’s activity, customer’s experience, acquisition strategies…).- Data visualisation and KPI monitoring (SQL/QlikView pipeline).- A/B testing of new product features (such as design, business related or algorithms parameters).- KPI design for new product features development
- AmazonTransportation AnalystLOGISTIQUE & SUPPLY CHAINmars 2017 - septembre 2017 (6 mois)Cité de Londres, Royaume-Uni- Project Management: KPI analytics to identify specific network weaknesses to reduce costs and volume of failures (late shipments, damaged shipments...).- Transportation data extraction (SQL) and visualisation (Python, Excel, Tableau)- Financial and Customer Service data analysis support.
Recommandations
Soyez le premier à recommander Matyas
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
- Diplôme d'Ingénieur en Sciences InformatiquesIMT Atlantique2017- Computer Sciences (OOP, Local Search, Graph Theory) - Optimization (Convex Optimization, Constraint Solving Problem) - Physical modelization (Robotics, Electrical Engineering)
- MSpe in Machine/Deep LearningTelecom ParisTechCourses: - Data Engineering (NoSQL / Distributed computation) - Statistical / Machine Learning - Deep Learning Projects: - Transfer Learning for a Question Answering bot using BERT pre-trained model (by Hugging Face) and fine-tuning on specific task and data. - Generation of classical music (J.S Bach alike) from MIDI files with a deep-RNN (LSTM layers) architecture model. - Construction of a NoSQL architecture to query over 90 GB data using Spark Scala for data pre-processing and MongoDB for queries (on AWS machines). - IoT challenge : predicting geo-location of connected devices based on incomplete GPS signals using models blending and leave-1 out cross validation. - Computer Vision challenge : finding the discriminative physical features of a labelised faces dataset using an handcrafted Inception Resnet model for classification, then extracting the relevant feature maps involved in the classifier's decision making.