Nada Staouite

Skills: Keras · Data Visualization · MapReduce · Deep Learning · Computer Science · Analytical Skills · Big Data · TensorFlow · Data Analysis · Artificial Intelligence (AI) · Cloud Computing · PySpark · Flask · Hadoop · SQL · Statistical Modeling · Databases · Docker Products · Object-Oriented Programming (OOP) · Data Science

Skills: Keras · Data Visualization · Deep Learning · Computer Science · Analytical Skills · TensorFlow · Artificial Intelligence (AI) · Cloud Computing · PySpark · SQL · Statistical Modeling · Object-Oriented Programming (OOP) · Data Science

Data quality audit Creation of BigQuery tables Customer segmentation analysis Churn prediction NLP model to predict the opening rate of newsletter emails

Deep Learning for Data to Text and Text to Data Generation: Adapting large‐scale Transformer models to language generation from structured data and data generation from textual data.

Data experimentation : ETL with Pyspark on Google Dataproc, creation of BigQuery Table, Machine learning model optimisation, MLflow traking , Interpretation ML pipelines : nodes sequencing with kedro , code linting , unit tests, MLflow Model Registry, REST API, application packaging with docker. Deploy model in production : CICD serverless pipelines, Kubernetes, Automation and planning with Airflow

Challenge : Yield optimization of a catalyst production line Solution : - Communication with the business managers about their expectations - Formulation of the business problem into a Machine Learning problem - Extraction and preprocessing of time series data coming from the platform sensors - Creating Machine Learning and Deep Learning models to predict the catalysts characteristics from the time series data - Extraction of the best operating parameters, delivery of the result to the core business

Challenge : Predict the type code of Rakuten France marketplace products as defined in their catalog. The data consists of texts and images of each products, and the goal is to deploy a multimodal deep learning model approach. Solution : - Text and image data analysis and preprocessing (noisy and imbalanced data) - Transformation of this large dataset into tensorflow binary records files to reduce the training time. - Building a multimodal deep learning model, with tensorflow, that takes images and texts data as input (NLP et Computer Vision encoders) , for a joint supervised classification training - Creation of a Streamlit web application to present the results

• Revenue optimization of a portfolio of routes , to guarantee the optimal income according to capacity and demand. • Performance analysis: Evolution of the main KPIs year over year, spilled and spoiled flights, trends of short and long term flights, high and low activity periods results. • Developing pricing strategy based on past results, competitor’s prices and events likely to influence demand… • Responsible for the flight capacity management: - Monthly adjustments of flight capacity based on profitabilty studies. - Coordinating these capacity changes with the scheduling department. • Monthly rolling forecast of aircraft load factor and average ticket price for the upcoming months. • Working with various departments such us: Network Planning, Flights Scheduling, Marketing, Tour Operators, Finance and Operations to provide them with feedback and figures to improve Transavia's results • Responsible for Revenue Management Tool: - Participating in calls with the provider and Transavia Netherlands Airline - Receiving the new releases - Organizing meetings to share the new features with the RM team that can improve our way of work. • Management of irregularities: choice of flights to be canceled in case of strike, operational or weather problems …

• Statutory audit of the financial corporate and consolidated statements of several companies such as Plastic Omnium, Alstom , Valeo, CNP assurance… • Development of audit strategy and evaluation of internal control procedures • Reconciliation of accounting data with management data. • Validation of financial statements, delivery of various attestations to our customers

• Statistical study to extract information from data collected through multiple surveys to improve of the bike paths of the city of Las Vegas. • Create Dashbords from databases provided by the Nevada Department of Transportation with Oracle BI Interactive Dashboards software…