Socio Data Management
Lead Data Scientist - Product Owner
Socio Data Management
Download technical file
PDF
Our consultant has over 12 years of experience during which he has built a solid background in data analysis, machine learning and agile methods for the development of Data Science solutions and platforms.During his career, he has designed and developed a large number of AI solutions in B2B and B2C. During his last work experiences, he was Lead Data Scientist, Product Owner and Scrum Master on several projects, using the latest technologies and frameworks in the field. Genuinely passionate about machine learning and custom solutions, our consultant has focused his latest efforts on NLP and Recommendation Engine, building robust and standalone solutions.

 

#DataAnalyse
#DataScience
#DataVisualisation
#BigData
#Statistics
#Econometrics
SKILLS
Python
Advanced - Certificated
Deep Learning - Computer Vision
Advanced - Certificated
NLP
Advanced - Certificated
Hadoop - Hive - Spark
Advanced
SQL Server - Kibana - Elasticsearch
Advanced
SAS - STATA - R
Advanced - Certificated
AWS Cloud Platform
Advanced
Product Owner - Scrum
Advanced - Certificated
Cybersecurity (ISO27001)
Advanced - Certificated
Risk Management (ISO27005)
Advanced - Certificated
LANGUAGES
TURKISH
Mother tongue
FRENCH
Advanced - Certificated
ENGLISH
Bilingual
PROFESSIONAL EXPERIENCES

2019 – Today

ARONDOR

Senior Data Scientist – Product Owner
Project: Product owner and SM: Management of “Product backlog” and the relationship with stakeholders, developer: definition of “Done”; coding and testing NLP logic
Python – WS – Elasticsearch – Kibana – SQL – NLP – Machine – Learning for anomaly detection – Java – AWS – Econometrics – SCRUM 

Issues:

Ensuring better customer churn.

Missions:

  • WS platform to perform real-time sentimental analysis of company emails with word processing and creating emotional and sentimental mapping of each customer to optimize customer service productivity.

Added value:

  • 450K Euros profit in 2019 and 2020
  • Estimated value of product is 1.5M for 2021

– – – –

Senior Data Scientist – Product Owner
Project: Product owner: Management of “Product backlog” and the relationship with stakeholders, developer: definition of “Done”, code and test textual analysis in object and clause detection
Python – Computer Vision – SaaS – Java – NLP – Machine Learning for classification – AWS – SCRUM

Issues:

Reducing data size and creating custom reports given client logs.

Missions:

  • Resizing of company documents: Module for the classification, decomposition, transformation of texts, graphs and tables and the generation of automatic summaries of large documents from user logs.

Added value:

  • ROI of the project is 14.5% in 2020 and 2021.

– – – –

Senior Data Scientist – Product Owner
Project: Product owner: Management of “Product backlog” and the relationship with stakeholders, developer: Definition of “Done”, coding and testing computer vision algorithms
Python – Spark – Computer Vision – Deep Learning – SaaS – Java – AWS – SCRUM

Issues:

Reduce time burden for video analyis by insurence agents.

Missions:

  • Computer vision module for the detection of events and accidents from sequential processing of moments and objects in videos of customer complaints.

Added value:

  • ROI of the project is 22.4% in 2020 and 2021.

– – – –

Senior Data Scientist – Product Owner
Project: Developer: Datawarehouse ETL and coding/testing algorithms
Python – Econometrics Applied microeconomy – WS – Server cloud – SCRUM

Issues:

Examine supplier failure risk, stock shortage, antitrust, corruption – fraud, cyber security risks.

Missions:

  • Development of the SaaS-IA interface for risk management in addition to the work of the Risks Managers to examine the risks of supplier failure, stock shortage, antitrust, corruption / fraud, cybersecurity risks.

Added value:

  • 510K euros profit in 2019-2020.

2018 / 11 months

AQUILA DATA ENABLER

Expert Data Scientist
Project: Consulting and development of AI solutions: coding data science-oriented algorithms, preparation of presentations and performing demos, planification and presentation of POC and MVP, managing agile projects: definition of “Done”, participating in scrum planning, creating multidisciplinary PhD team, technical interviews with data scientist candidates, founder member of the « AI&ML Lab »
Python – SQL – NLP – NLU – Traditional Machine Learning methods and Deep Learning – CNN, Hadoop – Linux – Econometrics – Tensorflow GPU – Darknet – OpenCV – Keras – Pandas – NLTK

Issues:

Creating custom AI solutions for each given context.

Missions:

  • Contract analysis: identification of signatures, dates, penalties, duration, etc. and data normalization.
  • Credit risk prediction: Company score prediction using data from ratings agencies and press articles.
  • Price prediction and recommendation: implementation of econometrics within deep learning models using customer/product pair data.
  • Gender and age prediction of website visitors.
  • Document compliance scoring from models of classification and identification.
  • Automatic processing of maintenance tickets (Calculation of the degree of criticality/priority, Search for root causes, Suggestions)
  • Extraction tables from PDF documents.
  • Customized recommendation engine for the energy and automotive sector among international clients.
  • Analysis of customer behavioral trends via web scraping and textual analysis.

Added value:

  • 270K euros profit in total.

2017 – 2018 / 23 months

DEEP MEANING DATA LABORATORY

Scientific projects manager in data science
Project: Management of « Product backlog », development: coding and testing NLP logic
Python – NLP : Word embedding – W2V – Text mining – Web scrapping

Issues:

Identify personal typology to select best candidates from CVs.

Missions:

  • Measurement of “Soft skills” via an emotional mapping of candidates and personality from CVs.

Added value:

  • 80K euros profit in total.

– – – –

Scientific projects manager in data science
Project: Research and implementation of statistical AI solutions, development, identification of metrics -KPIs- and validation: coding data science algorithms and performing technical tests, management of multidisciplinary and international projects (San Fransisco)
Python – Kafka – SCALA – Hadoop – R – SAS – STATA – SQL – NLP – Classification and clustering with Deep Learning – Recommendation Econometrics – Tensorflow – Theano – CNN

Issues:

  • Sentimental analaysis for fraud detections.
  • Creating better KPIs metrics and accuracy measurement.
  • Real time fraud detection.
  • Automatization International transactions control.

Missions:

  • Identification of abnormal behavior in financial e-mails (Antitrust).
  • Development of new statistical tests for the validation of classic Machine Learning and Deep Learning methods (convolutional neural network with TensorFlow, Theano).
  • Development of statistical models to analyze and to process complex behavioral data sets for fraud detection and better classification of false negative transactions giving 100% recall.
  • Multidisciplinary model (econometric and ML) specifications for the fraud detection in international transaction (SWIFT, SEPA-ISO20022) and credit card.

Added value:

  • 287K euros profit in total.

2011 – 2015 / 4 years

TRS Groups

Product Developer and Data Management
Project: Data Management, identification of metrics -KPIs- and validation, price prediction modeling of luxury products and product recommendation
Classification – Recommendation – Prediction of demand elasticities – Datawarehouse – STATA – R – Python – SQL.

Issues:

Identification of new products.

Missions:

  • Programming the deployment of the life cycle of new luxury products.

– – – –

Product Developer and Data Management
Project: Data Management, identification of metrics -KPIs- and validation, price prediction modeling of luxury products and product recommendation
Classification – Recommendation – Prediction of demand elasticities – Datawarehouse – STATA – R – Python – SQL

Issues:

Advanced data solutions using stitistical and economic models.

Missions:

  • Definition of data models, security and storage of sensitive personal data Extraction and transformation of raw data to better capture and analyze consumer trends.
  • Creation of advanced statistical models to control the price of based on pairs of the customer profile and goods.
  • Exploitation of customer data via econometric models to overcome the problem of cold start.

Added value:

  • 785K euros profit in total for both missions.

2003 – 2005 / 2 years

MINISTRY OF FINANCE

Consulting-platform SaaS
Project: Working in the tax automation and transformation process, identification of user scenarios, implementation of SaaS, operation of the tax automation system
C++ – Java – HTML – BackUp DATA – SaaS

Issues:

Digital transformation of tax system.

Missions:

  • Prepare use cases and scenarios and tests.
  • Operational troubleshooting in tax offices.
  • Ensure and save daily data.
  • Inform IT engineers about the taxation process.
  • Instruction and sensibilization of tax inspectors.
QUALIFICATIONS

2018

CNRS-CES Axe Theoretical Microeconomics

Postdoc in Econometrics and Data Science

2017

École Polytechnique X 

Master 2 in Data Science

2014 – 2017

University of Paris 1 Panthéon – Sorbonne

Ph.D in Applied Microeconomics with Data Science
Very honourable distinction with unanimous congratulations from the jury

2011 – 2014

University of Paris 1 Panthéon – Sorbonne

Ph.D in Economics

FOR MORE INFORMATION

CONTACT US