Fouad Amrani

Au sein du CISIRH (Centre Interministériel de Services Informatiques Ressources Humaines), participation à la transformation numérique des ressources humaines de l’État. Missions : • Evaluation des jobs en cours sur l'ETL existant (Talend) • Migration des jobs de Talend vers Python

Data Engineer consultant at SACEM (international copyright management company) Missions : - Creation of data pipeline currently in production - Creation of Datamarts according the business needs (SQL ou PySpark) - Creation of table on snowflake based on varying sources (parquets files, csv archives, varying databases, API) - Data gouvernance - Monitoring the proper execution of data pipelines already in production - Unit tests Cloud (AWS Services) : S3, EC2, Lambda, EMR clusters, RDS, CloudWatch, StepFunction Languages/Framework : Bash, Python, PySpark, SQL Outils : Linux, AWS CLI, Terraform, Git, GitLab

Definition of need in partnership with buisness (also with the Data team) Création d'un modèle de reconnaissance de véhicules (YOLO) Creation of a machine learning model to recognize vehicles (YOLO) Labellisation et entraînement du modèle Data labelization and model training Participation to putting into production on servers presents construction sites then on the cloud (AWS)

Queries via the SQL language on the SNDS (formerly SNIIRAM), which is one of the largest medico-administrative databases in the world. Impact measurement general practitioner investigation campaign affixing «Non-substituable». Participated in the drafting of briefing notes. Updated reports on Drug Reimbursement Amounts.

Collection of environmental data from Accredited Agencies for Air Quality Monitoring (AASQA). Statistical analysis within the European project MED'HISS carried out on 8375 subjects. Preparation (writing) of original scientific articles in peer-reviewed international journals. Presentation of a poster at the International Society for Environmental Epidemiology (International Conference on Epidemiology) Rome, Italy (1-5 September 2016).

Participation in the ASTHM' CHILD project: - Identify the main sources and determinants of various pollutants within the housing (allergens, phthalates, terpenoids, synthetic musks...) - Construction of predictive models to predict the level of pollutant concentration using predictors: housing characteristics and habits of residents within dwellings. To answer these questions we use methods from Data Mining (machine learning) Software used: RapidMiner, Weka, SAS