Ahmed CHEIKH ROUHOU

DE: - Develop ingestion tools from differents flows (csv, xml, xlsx) - ETL: Insertion in Databases. - ETL: KPI jobs orchestration. - ETL: Pseudonymisation and Anonymisation of sensitive data. - Infrastructure deployment for different usecases in GCP. - Database synchronisation (Singer IO) Skills: Python, Airflow, AWS, Jenkins, Terraform DS: - Supervised: Classification model creation to predict fields in invoices - API deployment to return the results of the model - Unsupervised: Clustering for the clients segmentation using transactionnal data (purchases) and analytics data (GA) Skills: GCP, Pandas, Numpy, Scikit Learn

- Develop algorithms for calculating risk parameters on investment funds - Develop fund categorization algorithms based on their type of portfolio, sector, size and investment factor using ML. - Maintain the database on AWS RDS (PostgreSQL). - Develop calculation batches in order to automate the launch of various calculation and data processing based on Scrapy and ActiveMQ.

- Modeling a problem of pre-selection of CVs in order to lighten the work of the recruiters. The goal is to calculate for a given offer a score per CV based on several weighted criteria. - Extraction of text streams of CVs from recruitment platforms and the corresponding offer using textract. - Creation and cleaning (stopwords) of a dictionary of words with the library nltk. - Transforming words into vector with the Gensim.word2vec library. - Creating word culsters using K-Means with the SKlearn library. - Development of a matching score calculation algorithm between a CV and a given offer. - Recovery from the text stream of other parameters (level of studies, areas of previous work experience). - Calculation of the final score and recovery of the best CVs.

It is a mission in the application support teams (functional and technical) of the risk perimeter of CACIB. Risk applications provide users (Risk & Permanent Control, Market Risk Management, CASA) with all types of risk indicators. Risk systems also cover rating applications. Global support uses ITIL's process approach and the LEAN approach to strive for operational excellence. At first I was trained in the business of risk application support to understand the business issues of this function. I then read the specifications set up and completed with the members of the Global Support team. The goal of the project is to automate the different support processes, I made the design of the different entities around the business and then implemented. The technologies used to lead this project: - JAVA / J2EE with the different frameworks needed (Hibernate, JasperReport) -HTML, CSS, JSP -SQL I worked in coordination with: - The members of the support team (Paris and Singapore ); - Project teams (MOE / MOA) (Paris and Singapore); - Production teams (ITS) (Paris and Singapore). I had to constantly ensure the respect of the standards in terms of development tools as well as the security procedures in force.

- Simulation of business creation in a multidisciplinary group consisting of four engineers, a manager and a designer. - Exchange with experts from BNP Paribas, Orange, Nokia and Accenture stakeholders, Boeing Air Defense. - The idea is to develop a game for children that allows them to discover the fauna of their region. The game consists of browsing the city in search of virtual animals in the manner of Pokemon Go through a radar, switch to interaction mode with the animal (heal, feed, clean the environment) and unlock content around the saved animal in question. - To promote the product, we have developed a prototype as proof of feasibility on Unity 3D and a promotional video with After Effect - The idea was presented during an afterwork networking with French Tech Mexico and the Embassy of France and during a presentation to a jury of innovation professionals at Numa Mexico (accelerator and incubator).

- Development of a GUI application using the tkinter and sqlite libraries on Python for managing a repository