Rihab B.

• Designed and implemented data pipelines for both batch and stream processing, optimizing data flow and efficiency;
• Explored and implemented data pipelines using AWS Glue and PySpark, ensuring scalability and robustness;
• Integrated Delta Lake into the pipelines to enable delta processing, enhancing data management capabilities;
• Developed job templating using Jinja to streamline the creation and management of data processing jobs;
• Built and automated data validation pipelines, ensuring the accuracy and reliability of processed data;
• Deployed and configured Trino to facilitate efficient data access and querying across various sources;
• Prepared comprehensive documentation for each component and tool explored, ensuring knowledge transfer and easy maintenance;
• Utilized tools such as Python, PySpark, Glue (Jobs, Crawlers, Catalogs), Athena, AWS, MWAA (Airflow), Kubernetes, Trino, and Jinja to achieve project goals.

Design & Implementation of a Forecasting Platform – Engie (French Global Energy Company)

• Designed and implemented a comprehensive forecasting platform tailored to the global energy sector.
• Developed data pipelines using Python and PySpark, ensuring efficient and scalable data processing.
• Orchestrated job workflows using Airflow and Databricks, optimizing task management and execution.
• Implemented data engineering processes utilizing Databricks’ Delta Live Tables (DLT) for robust data management.
• Built and deployed data stream processing pipelines using DLTs, enabling real-time data processing capabilities.
• Developed Feature Store APIs for interaction with components and created reusable templates to standardize processes.
• Utilized MLflow to build, manage, and track experiments and machine learning models, ensuring rigorous experimentation.
• Managed the lifecycle of ML models using MLOps techniques, implementing reusable templates for consistency and efficiency.
• Created dashboards for data analysis and visualization, facilitating data-driven decision-making.
• Developed APIs using .NET/C# to expose data, ensuring seamless integration and accessibility across systems.
• Employed tools such as Databricks, PySpark, Python, R, SQL, Glue, Athena, Kubernetes, and Airflow to deliver a robust and scalable solution.

• Led projects focused on integrating Large Language Models (LLM) and AI technologies, driving innovation within the organization;
• Assisted in designing and implementing data migration solutions, ensuring seamless transitions for various clients;
• Developed integrations and clients for vector databases, leveraging different open-source AI tools to enhance capabilities;
• Actively communicated with clients to gather requirements and ensure alignment with their specific needs;
• Utilized tools such as Python, Google Cloud Platform (GCP), and Datastax to deliver robust solutions.

Building and supporting promotion planning demo solution

• Developed generic data pipelines to transform raw client data into a format compatible with the data model of the promotion planning demo system;

• Wrote scripts to generate meaningful business data, ensuring alignment with the needs of the application;

• Collaborated with the science team to understand business requirements and determine the necessary data transformations to enhance data utility;

• Designed and implemented a generic PySpark codebase that efficiently transforms data to fit the required data model;

• Utilized tools such as PySpark, JupyterHub, Kubernetes, and Azure Data Lake to execute and support the project.

Implementing and Migrating Data Pipelines, and Supporting Legacy Systems - SumUp (Fintech German Company)

• Designed and implemented data pipelines for both batch and stream processing, optimizing data flow and efficiency;

• Explored and implemented data pipelines using AWS Glue and PySpark, ensuring scalability and robustness;

• Integrated Delta Lake into the pipelines to enable delta processing, enhancing data management capabilities;

• Developed job templating using Jinja to streamline the creation and management of data processing jobs;

• Built and automated data validation pipelines, ensuring the accuracy and reliability of processed data;

• Deployed and configured Trino to facilitate efficient data access and querying across various sources;

• Prepared comprehensive documentation for each component and tool explored, ensuring knowledge transfer and easy maintenance;

• Utilized tools such as Python, PySpark, Glue (Jobs, Crawlers, Catalogs), Athena, AWS, MWAA (Airflow), Kubernetes, Trino, and Jinja to achieve project goals.

Building a Data Curation Platform

• Implemented a platform designed to make building data pipelines generic, easy, scalable, and quick to assemble for any new client;

• Prepared detailed design documents, architectural blueprints, and specifications for the platform;

• Gathered and documented requirements, creating specific epics and tasks, and efficiently distributed work among team members;

• Developed command-line and pipeline functionalities that enable chaining transformations, facilitating the creation of generic data pipelines;

• Supported the management of metadata for various entities defined within the platform;

• Conducted runtime analysis and optimized the performance of different platform functionalities;

• Studied scalability requirements and designed performance improvement strategies to enhance the platform's robustness;

• Built a PySpark interface to facilitate seamless integration with data science workflows.

• Developed generic data pipelines to transform raw client data into a format compatible with the data model of the promotion planning demo system;
• Wrote scripts to generate meaningful business data, ensuring alignment with the needs of the application;
• Collaborated with the science team to understand business requirements and determine the necessary data transformations to enhance data utility;
• Designed and implemented a generic PySpark codebase that efficiently transforms data to fit the required data model;
• Utilized tools such as PySpark, JupyterHub, Kubernetes, and Azure Data Lake to execute and support the project.

• Led the team in building data pipelines to support a retailer's promotion planning solution;
• Participated in meetings with business and data science teams to understand and identify project needs;
• Collaborated with the team to translate business requirements into actionable epics and stories;
• Designed and implemented the identified business requirements, ensuring alignment with project goals;
• Developed and executed unit tests to ensure the functional correctness of implementations;
• Created a data loader application using Scala Spark to load data from Parquet files into Cosmos DB/Cassandra API;
• Implemented an online forecaster API using Scala, Akka, and Docker to enable real-time promotion forecasting;
• Managed the deployment of the project on the client’s Kubernetes cluster, ensuring smooth operation and integration;
• Utilized tools such as Scala, Spark, Azure Databricks, Azure Data Lake, and Kubernetes to achieve project objectives.

Project 1: Building a Speech Recognition Solution

• Developed a speech recognition solution aimed at transforming retailers' questions and commands into actionable tasks executed against a user interface (UI);

• Utilized TensorFlow, Python, AWS, and Node.js to design and implement the solution, ensuring seamless interaction between the speech recognition engine and the UI.

Project 2: Design and Implementation of a Short Life Cycle Forecasting System

• Prepared comprehensive design documents and conducted studies on existing AI solutions, with a focus on voice and speech recognition capabilities;

• Collaborated with the team to prepare and collect relevant data for the project;

• Executed the processes of data augmentation, validation, and transformation to extract essential information for forecasting purposes;

• Contributed to building a user interface and integrated backend functionalities using tools such as TensorFlow, Python, AWS, JavaScript, Node.js, Scala, and Spark.

• Designed and structured the architecture for various components of a retail forecasting project;

• Implemented and deployed key components, ensuring seamless functionality within the overall system;

• Integrated all components, automating the processes and establishing an end-to-end batch process for streamlined operations;

• Optimized the runtime and performance of each component, enhancing the system's overall efficiency;

• Developed forecast comparison templates to facilitate the evaluation of forecast quality, aiding in accurate performance assessments;

• Utilized Logicblox, Python, and Tableau Software to achieve project goals, ensuring high-quality results.