Nassim EL GHAZAZ

Context - Arkema HIP (Hybrid Integration Platform) Datalake Project Lead: - Design and Implement a datalake to manage the company's various domains (sales, authorizations, quality, environment-health-safety, master data, etc.) while meeting internal application needs. - Design and Prepare S3 buckets: Landing – Raw – Standardized – Published, on 3 environments: Dev, No-Prod and Prod. - Create data glue catalog containing several databases for each domain of the company. - Implement Serverless ETL & ELT data flows with AWS services Glue, Athena, Lambda and StepFunctions, which are triggered by file drop events on the Landing Bucket or on scheduled events. - Implement data integration tests associated with data pipelines. - Implement Near Real Time flows for on-the-fly data processing using Kinesis Firehose, Lambda and SQS services to create a Sales Datalake. - Verify the received files and ensure compliance with the Datalake spec, clean and aggregate data according to different use cases. • Communicate with internal Data Scientists from R&D team, to prepare datasets accordingly to their needs, implementing data quality rules, and applying data profiling using SageMaker Python Notebooks. • Publishing data feeds to internal databases from the published layer of the datalake (files publishing and/ or POST requests, ODBC connection with an RDS etc) and authorizing internal clients to consume data by scope (Power BI, IBM analytics, etc.). • Create the CloudFormation YAML templates and deployment scripts / Infrastructure as Code (SAM & CDK) - Gitlab CI/ CD chain via automated pipelines with stages: unit tests, code scan, linting, data quality, integration tests, deploy. Technical Environment: • AWS (S3, Lambda, StepFunctions, Kinesis, API Gateway, CloudFormation, Glue, Athena, Lakeformation, ECR, ECS, SQS, DynamoDB, RDS, SageMaker, CloudWatch …) • Python, TypeScript, ShellScript • GitLab, Datadog, Veracode

Sectors of activity : Energy, Highway, Agriculture, Health, Tourism, Banking. • Design Task Flows made of Talend jobs to migrate data from a NoSQL CosmosDB Data Lake to an Oracle database and vice versa. • Carry out the deployment of Big Data jobs executables on Microsoft Azure and automation using WebJobs. • Set up a Data Gateway connection to synchronize a Legacy system with an Azure Data Lake. • Create an innovative platform “Mobility as a Sustainable Service” (MaaSS) made up of two components (Web Portal and Hybrid Mobile App), and building algorithms that predict the type of vehicule that corresponds to fluctuations of smartphone sensors data. • Data mining & big data audit on production chain and highway tolls. • Creation of an augmented reality application for an Urban Farm : The main functionalities of this android App are : Retrieving data from an external Api, Showing KPIs to Users, and the next tasks to perform related to plants cultivation by showing augmented objects in the physical world. • Carry out a state of the art of predictive maintenance in the context of Industry 4.0. • Perform predictive maintenance workflow tasks: - Data Processing and evaluation of Machine Learning models for breakdowns prediction inside the urban farm materials. - Data Visualization / Reporting of results related to predictive maintenance. • Predictive analysis of time series on behalf of a French energy distributor. • Text mining / natural language processing, OCR and extraction of informations from pharmaceutical documents. - Environment : Python, Spark, C#, Tensorflow, Google Cloud Platform, Microsoft Azure, Talend Big Data Platform, Dash Plotly, Hadoop, Kafka, Hive, Grafana, Docker, Google Arcore, git, Jenkins, Docker, Web and Mobile technologies (Angular, Node Js, Ionic & Cordova Frameworks), DB Management (SQL and NoSQL).

Tour Operator – Data Science project: • Clean and structure customer data. • Carry out statistical analyzes (Descriptive, Inferential and Predictive). • Apply Machine Learning algorithms to forecast time series of overnight stays and turnover on several sites (vacation villages). • Restitution of prediction results on a dashboard integrated in a web application (Data Visualization). Retirement Provident – Applied Mathematics project: • Solve optimization problems to offer the most optimal annuity scenarios for clients of the retirement provident division. • Present the results in Excel file format made up of interactive sheets (Buttons that launch macros).

• Modeling of mathematical problems of assigning logistic agents to tasks related to inventory, equipment movement and bio-cleaning (NP-Hard problem). • Realization of a Back Office Mission solution, which implements the predictive method of resources allocation and which allows the management of hospital equipment. - Meta-heuristic resolution method used : Genetic Algorithms. (JAVA REST, Javascript)

• Analyze the existing data platform and collect relevant indicators (KPI concerning road control and training of professional drivers). • Design and Implementation of a Datawarehouse (MSSQL) and supply it with data from the SITR Road Transport Information System (MS BI - SSIS). • Preparation of reports relating to each module via the data warehouse (MS BI - SSAS and SSRS) and creation of a dashboard designed for the national observatory of transport and logistics. • Time series analysis on data related to training of professional drivers and road control. Generation of forecasts after the evaluation of different statistical models (R). - Participation in the Design and creation of a web portal (CMS Wordpress, HTML5).

• Analyze the existing workflow and collect data related to each department. • Study and analysis of the business processes of the information systems and freight logistics department. • Design of a relational database (MSSQL) and creation of management application for the provisional maintenance calendar (C # .NET).