Ismail Mezzour

DBT & Airflow - Contributed to the setup and standardization of dbt across Accor’s data ecosystem - Set up and maintained Apache Airflow on Astronomer - Built reusable Airflow templates to generate dbt DAGs - Supported data teams with best practices on orchestration and scheduling - Wrote onboarding documentation for dbt and Airflow - Orchestrated Snowflake stored procedures and tasks - Implemented monitoring DAGs for reliability CI/CD for Data Pipeline Deployments - Implemented GitLab CI/CD for dbt and Airflow deployments - Set up multi-stage deployments (dev, pre-prod, prod) - Designed tag-based configurations to route teams through CI/CD pipelines and reduce duplication - Added unit tests and automated checks - Reduced Astronomer deployment time from ~5 minutes to ~1 minute - Configured pre-commit hooks and sqlfmt Snowflake - Built FinOps dashboards to monitor and optimize warehouse costs - Managed Snowflake infrastructure as code using Terraform - Set up a pre-production environment for dbt testing - Audited access and usage logs and built Snowsight dashboards - Optimized warehouse sizing, auto-suspend, and workload isolation - Defined role-based access models and governance - Benchmarked semantic layer solutions Data Quality & Observability - Implemented dbt tests for data integrity - Set up Soda quality checks - Benchmarked observability tools (Sifflet, Monte Carlo) - Delivered POCs with Sifflet, Elementary, and Soda ML Ops - Delivered POCs for model registries (Snowflake, SageMaker, GitLab) - Worked with data scientists to optimize ML workflows - Introduced core data science concepts - Set up ML model monitoring in Snowflake AWS - Managed AWS infrastructure as code using Terraform - Built Terraform modules for scheduling and monitoring dbt - Developed AWS Lambda functions to send data from Snowflake to Splunk

ETL and Data Pipeline Development - Develop and maintain ETL jobs with Python, Spark, SQL, DBT, and Airflow - Develop tools to improve data quality monitoring - Model development - Optimization/profiling of Spark jobs - Set up cloud functions to transform the data - Set up a POC on data flow Cloud Platform Management - Participate in the design, maintenance, and documentation of Google Warehouse BigQuery - Set up Microsoft Azure workbooks to follow the night batch pipeline metrics - Working on Microsoft Azure and Google Cloud platforms Collaboration and Project Management - Interact with business teams to understand their data needs - Collaborate with the Data Science and Data Analysis teams to further develop a data platform - Functional and technical POC - Technical framing to estimate the workload of a project - Working on the scrum framework - Animation of the daily scrum - Set up a report/summary on what we did in the previous sprint and what we will do in the next sprint

- Reporting and monitoring with SAP Analytics cloud - Creation of dashboards to allow teams to have information on their products in real time and with a high level of quality. - Data engineering with python and sql - Setting up a connection between the SAP Analytics cloud and an Odata service through a cloud connector. - Creation of an executable for data processing (python)

Project made in python: - Logstash and grafana data recovery - Data processing - Time series prediction using recurrent neural networks (LSTM, GRU, ARIMA) - System monitoring based on the predictions we have made.

Automation of tasks: project carried out in VBA - recovery of data from salesman files