Lahcen Seddir

– Designed and developed dbt models on Snowflake, following best practices for modularity, documentation, and maintainability. – Built reusable dbt macros to automate complex business rules and ensure scalable logic implementation. – Implemented unit, data quality, and performance tests using dbt and Elementary to ensure robustness and reliability. – Enabled end-to-end data lineage in dbt, providing full visibility from raw ingestion to reporting layers. – Developed a Snowpark-based scoring program to support lead generation use cases, including churn detection and dormant lead reactivation. – Orchestrated end-to-end workflows using Apache Airflow, automating dbt runs, testing, and reporting processes. – Automated data processing with Python and integrated external APIs into dbt-driven pipelines. – Collaborated closely with product, data, and business teams in a Scrum environment to align technical solutions with business needs. – Monitored pipeline and model performance and proactively proposed continuous improvements. – Deployed and operated solutions in a cloud environment (Azure), handling authentication, security, and resource configuration.

– Designed a Star Schema data warehouse in collaboration with business stakeholders to optimize analytical performance. – Developed complex PySpark ETL/ELT pipelines sourcing data from SQL/NoSQL databases, flat files, and third-party APIs. – Implemented real-time streaming pipelines using Cloudera DataFlow (NiFi) for IoT data ingestion. – Orchestrated end-to-end workflows with CDE Airflow, including dependency management, retries, and error handling. – Analyzed and optimized Spark jobs to address performance bottlenecks. – Applied data management best practices (Parquet/ORC formats, compression) to optimize execution time and storage usage. – Developed unit and integration tests to validate data transformations. – Implemented CI/CD pipelines to ensure reliable deployment of Spark workloads. – Migrated PySpark jobs from Databricks to Cloudera CDE, adapting dependencies, storage paths, and Spark configurations. – Results: 60% improvement in ETL pipeline performance following Spark and storage optimizations. – Results: 20% reduction in storage costs through compression and retention strategies.

– Designed and optimized ETL/ELT pipelines for data ingestion from SQL and NoSQL sources. – Developed Spark Scala processing jobs to address complex business requirements. – Built REST APIs using Node.js and FastAPI to securely expose processed data. – Optimized Spark workloads through partitioning, memory tuning, and parallelism. – Supported migration to Google Cloud Platform (GCP), including data storage in GCS, loading into BigQuery, SQL-based modeling, and data quality validation. – Results: 50% reduction in Spark job execution time. – Results: Approximately 25% reduction in cluster costs through resource optimization. – Results: Secure and progressive migration of legacy workflows to GCP with no business disruption.

– Designed and implemented machine learning models for churn prediction. – Built a social listening platform to analyze and monitor brand mentions across social networks. – Developed NLP solutions for CV parsing and candidate-job matching using deep learning techniques. – Implemented ETL pipelines using Talend and Python to ingest data from ERP and CRM systems into data lakes. – Administered the Hadoop ecosystem and implemented data governance using Apache Ranger and Apache Atlas. – Designed and maintained an on-premise Cloudera Lakehouse architecture. – Results: 30% reduction in storage costs and 10x improvement in query performance after platform optimization. – Results: 80% reduction in operational workload through pipeline automation. – Results: Significant improvement in data quality and reliability for decision-making processes.