Zaradacht Taifour

◼ The project involved the build and run of Lacoste's data platform on Google Cloud Platform (GCP). I worked on both the functional side (directly engaging business teams to understand requirements and translate them into code) and the technical side (building and maintaining the data infrastructure).

◼ With a FinOps-oriented approach, I successfully reduced Lacoste's GCP costs by 30% from $10,500/month to $6,500/month, through a series of performance and cost optimizations that I designed and implemented.

SCOPE OF WORK:

◼ Reduced GCP costs by 30% (from $10.5K → $6.5K per month) through BigQuery optimization, resource rightsizing, and data lifecycle improvements.

◼ Acted as the main point of contact with the Meta team.

◼ Managed the project from scratch, covering project scoping, functional alignment, data modeling, implementation, and monitoring.

◼ Designed and implemented data ingestion pipelines within the marketing and retail analytics domain, ensuring high reliability, scalability, and data governance.

◼ Built end-to-end ingestion workflows from Google Ads Data Hub to BigQuery, securing data integrity and compliance with privacy standards.

◼ Implemented automated data backup and recovery mechanisms at the code level to ensure business continuity and resilience.

◼ Deployed and managed applications on Artifact Registry and Google Kubernetes Engine (GKE), optimizing for scalability, performance, and cost efficiency.

◼ Developed and maintained marketing connectors and API bridges to synchronize data across BigQuery, Snowflake, GA4, Meta CAPI, and FTP/SFTP sources.

◼ Led the end-to-end implementation of the Meta Offline Conversions pipeline, enabling automatic daily ingestion of purchase events to Meta Ads.

◼ Orchestrated dbt transformations, data quality checks, and audience activation pipelines, ensuring data reliability across marketing use cases.

SCOPE OF WORK: ◼ Designed and implemented the initial data platform solution to enable the transition of insurance data workloads to Microsoft Azure.

◼ Built the data mediation layer serving as the bridge between on-premises systems and the Azure cloud environment, ensuring seamless synchronization and reliability.

◼ Developed a data anonymization and security layer to protect sensitive information before cloud ingestion, including encryption and controlled data access mechanisms.

◼ Designed and deployed a three-tier data architecture (Bronze / Silver / Gold) to structure and manage data flows efficiently across the Azure ecosystem.

◼ Implemented data transformation pipelines in alignment with the medallion architecture principles, improving clarity and maintainability across processing stages.

◼ Acted as the main interlocutor between the engineering and data science teams, ensuring that data was properly prepared, documented, and delivered for model training and evaluation.

◼ Focused primarily on the cloud infrastructure and data mediation layer, ensuring performance, governance, and scalability of the overall solution.

SCOPE OF WORK: ◼ Designed and maintained data ingestion pipelines from Cloud Storage to BigQuery.

◼ Industrialized new ingestion flows via SFTP, Cloud Storage, and Kubernetes for scalable automation.

◼ Implemented and maintained dbt transformation pipelines, ensuring consistency and reliability of analytical models.

◼ Centralized data from multiple sources Snowflake, Google Analytics, and Cloud Storage into BigQuery for unified reporting.

◼ Monitored and maintained the Kubernetes infrastructure to ensure performance, stability, and cost efficiency.

◼ Developed marketing connectors to push customer audiences and campaign data to Meta (Facebook Ads), Criteo, and Google Ads.

◼ Configured and managed Airbyte ingestion workflows, enabling smooth integration between Snowflake and BigQuery environments.

TECHNICAL TOOLS & SKILLS: ◼ GCP, BigQuery, dbt, Kubernetes, Airbyte, Snowflake, Cloud Storage, GA4, Facebook Ads, Google Ads, Criteo