Freelancer profile translated to English.

Description

I have worked for 10 years on IT projects as a project manager and functional consultant. I was able to learn a lot, but something was missing: I didn't feel like I was part of the technological revolution that is happening. That's why I decided to undertake a specialized master's degree in Data Science / Big Data at Télécom ParisTech.

The main objectives were:

- to acquire the methods for working effectively and relevantly with data using Python

- to gain in-depth knowledge in statistics, machine learning, and deep learning

- to be able to use the latest Big Data technologies (Hadoop / Spark / NoSQL)

I was able to carry out several professional projects (tweet analysis for SFR, image recognition for Carrefour, automatic invoice recognition for Dhatim) as well as personal projects (prediction of football match results, centralized display of apartment listings from different websites...).

I therefore now have both the organizational and technical skills to successfully complete your Data Science projects and ensure their success.

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

Dhatim
Data scientist
TECH
February 2018 - Today (8 years and 4 months)
Paris, France
Responsible for the automatic invoice recognition process.

- Industrialization of the machine learning environment:
+ Definition of environments and workflows for labeling and training models
+ Implementation of the feature generation pipeline, model training, and performance metric calculation

- Definition of train and test datasets, feature engineering, tuning, metric definition

- Implementation of various Machine Learning / Deep Learning models to extract information from invoices, classify documents (invoice, bank statement, tax return, etc.), and determine if an invoice is a purchase or sale.

- Image processing:
+ Grid and column detection
+ Deep learning-based zone detection
Data science Machine learning Python TensorFlow Scikit-learn Pandas
Projets Personnels
Data Scientist
TECH
January 2018 - February 2018 (2 months)
Paris, France
- Creation of a football match result prediction model (web crawling, data modeling, feature engineering, machine learning)

- Participation in the Kaggle competition for click-through rate (CTR) prediction for an advertising company (over 250 million records)

- Analysis of housing advertisements in the canton of Zurich and creation of a model to predict the most interesting offers (web crawling, machine learning with Spark, data visualization with Tableau)
Data science Machine learning Hadoop Spark Tableau software Python Apache Spark MLlib
carrefour
Data Scientist
TECH
July 2017 - December 2017 (6 months)
Paris, France
Development in Python of an application for detecting anomalies and identifying products from images of supermarket shelves.
- Definition of the solution using machine learning and deep learning algorithms
- Development of an annotation tool
- Data preparation
- Model construction and performance measurement
- Writing of functional and technical documentation
Data science Machine learning Python TensorFlow Scikit-learn Pandas Data visualisation Data mining JavaScript HTML5 CSS3 Git Atlassian JIRA Atlassian Confluence Microsoft Office