Fabien Gadet

- Built end-to-end data platform from scratch using DBT (6-month initial development, ongoing feature expansion)

- Independently developed all data/DBT architecture and pipelines

Consolidated 24 second-hand watch marketplaces into unified analytics system.

- Engineered ETL pipelines standardizing diverse business models (auctions, subscriptions, individual listings) into single data model

- Implemented fuzzy matching algorithms with multi-stage logic to handle partial/incorrect reference numbers using watch features

- Automated weekly incremental data ingestion with 50+ validation tests and data freshness alerts

- Built comprehensive table documentation covering all logic and data lineage

- Created custom tool for selective table refresh when needed

- Built global MSRP tracking with multi-currency conversion based on client geography/demand

- Processed 4M+ product images weekly using RabbitMQ across all projects

- Delivered Tableau dashboards with custom index calculations for well-known luxury goods companies

- Created configurable email automation using Airtable and Tableau APIs with product-specific templates currently running 10 emails weekly/monthly/quarterly across 4 projects.

- Provided on-demand custom exports (typical 2-3 week implementation cycles)

- Platform success resulted in 2 major client acquisitions with 2 additional prospects in pipeline

- Migrated legacy Airflow and PostgreSQL from 2017 versions to current releases

- Implemented CI/CD pipelines using GitLab runners, Docker registries, and GitLab Pages

- Automated S3 backups and data lifecycle management with purging for oversized tables

- Integrated GitLab-Airtable automation workflows

- Introduced automation tools (n8n) and containerization standards

- All systems orchestrated through Apache Airflow with comprehensive monitoring and alerting

I was the first new data engineer to join after the Docker split (Docker sold Swarm and was rebranded with 60 employees) of the entire data environment. I helped recruit 7 data engineers and 1 data analyst. I migrated legacy code, led POCs to modernize the stack, and was in charge of Looker (data visualization tool). I spent the last 6 months working extensively on Snowflake administration.

Data visualization

I was in charge of Looker the first year, I provided quick introductions to every new employee that needed it and reviewed/merged ALL the PRs.

Full migration + improvement of our legacy code (Spark + Python / Jenkins)

Built the core pipelines of DBT that became the main pipelines afterward.

Interviews

I conducted around 20 interviews, mostly for data engineers (including 7 who were hired afterward) / data analysts, and data scientists.

- MonteCarlo

- DBT (approved)

- BigEye

- Dagster (approved)

- Alembic (approved)

- Atlan (approved)

- Snowflake administration

I spent the last 6 months administrating Snowflake with a colleague, which included:

- Full rework of the role system with versioning.

- Scripts that run hourly to alert/detect data freshness, Snowpipe errors, etc.

- Scripts for credit consumption that alert on spikes and provide a daily report of credit usage per warehouse...

- Scripts that detect heavy queries (mostly from Looker) that are costly and can be improved.

Team reinforcement to help release a major project in PySpark / Databricks