Zineb Lamrani

Context: Design and implementation of solutions for the analysis of data relating to different types of business financing

• Modeling and development of a generic ETL framework based on spark which allows to produce datasets, deploy ETL pipelines via AWS Glue to read, transform and write data using spark from different sources (s3, mongodb, athena, kafka etc ...) to different targets (s3, mongodb, athena, kafka etc ...)

• Templating of applications (via cookiecutter) to industrialize and facilitate the use of the framework by all teams.

• Creation of data pipeline for the different businesses (guarantee, financing ...)

• Optimization of data flows

• Architecture documentation

• Analysis and calculation of KPIs for the different businesses (financing, guarantee, innovation, digital…)

• Training and support of junior developers

• Writing user manuals

• Production of applications and operational maintenance

Technical environment: Python, Spark, Airflow, AWS (Athena, S3, Glue), Qlik Sense, Bash, Jupyter, Dataiku, Gitlab, Jenkins, SonarQube,

Datadog Pilot tools: Jira

Context: Implementation of a big data platform for the prediction, analysis of flows and breakdowns of "RER A" trains for the "Railway rolling stock" department:

• Extraction, transformation and analysis of complex data sets from multiple sources using tools such as Python, Spark

• Modeling, implementation and maintenance of datasets of rolling stock attributes (prognostics, traffic, maintenance, functions…)

• Setting up architecture and installing Qlik Sense server in multi-node mode

• Management of specific load balancing rules across all servers

• Development of automation scripts for the management of continuous reloading of business applications

• Analysis and calculation of KPIs for identifying precursors of failures with key users

• Development of dashboards for managing flows, delays and breakdowns of trains in real time,

• Development, evolution and maintenance of extensions for specific KPI calculations

• Management and facilitation of workshops with Key Users

• Implementation of the data dictionary

• Writing technical and functional specifications

• Training and support of users of different levels for the realization of their own Data reports

Context: Implementation of a trend analysis and anticipation solution based on Deep Learning algorithms for the luxury, fashion and cosmetics sectors. Achievements:

• Writing detailed technical and functional specifications

• Design and implementation of data project with construction of the collection, processing and storage pipeline process,

• Modeling and management of datasets (attributes, images, consumer comments, product information ...),

• Image segmentation via deep learning models, sentiment analysis from consumer comments,

• Management and facilitation of workshops with Key Users (stylists, trend specialists, influencers ...),

• Development of product dashboards (product information, price, description ...),

• Development and implementation of specific dashboards for on-demand reports for designers, marketing and sales managers (digital marketing KPIs, analysis of competitor prices ...) Technical environment: PostgreSQL, Python, Qlik sense, AWS, Scikit-Learn, Jupyter, Gitlab. Pilot tools: Jira