Name: Data & AI Tech Lead
Price: 750 EUR
Availability: InStock

Freelancer profile translated to English.

Back to original language

Languages

French
Native or bilingual
English
Native or bilingual
Arabic
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 50km)

Möet Hennessy,
Data & AI Tech Lead
LUXURY GOODS
January 2026 - June 2026 (5 months)
Épernay, France
Embedded within Moët Hennessy's champagne team, this mission focuses on leading
the data engineering and AI practice, bridging business strategy with modern cloud-
native architectures. It encompasses the design, build, and governance of the
maison's data platform on Google Cloud, ensuring every pipeline, model, and
dashboard delivers measurable value to the business.
Designing and implementing scalable data architectures on BigQuery, leveraging dbt for transformation layers that serve analytics and operational use cases across the organization.
Driving FinOps governance for BigQuery — optimizing costs, enforcing slot and storage policies, and providing spend visibility to stakeholders.
Building and deploying machine learning models on Vertex AI to address high-impact business needs such as harvest forecasting, and energy consumption anomalies.
Engineering robust, reproducible cloud infrastructure using Terraform, following Infrastructure-as-Code best practices.
Owning the CI/CD pipeline strategy on GitLab, ensuring smooth, reliable
deployments across all environments.
Reviewing and validating merge requests to uphold code quality, consistency, and architectural standards across the team.
Mentoring team members on day-to-day technical challenges across the full stack.
Monitoring production runs and proactively troubleshooting incidents to
guarantee platform reliability.
Technical Stack: Google BigQuery, Vertex AI, Terraform, DBT, Apache Airflow, SQL, Python, Pub/Sub, GitLab.
Big Query Google cloud Data architecture Data Platform Machine learning
Veolia North America
Cloud Data Engineer
ENERGY AND UTILITIES
December 2023 - December 2025 (2 years)
New Jersey, USA
As part of a strategic data initiative, the mission focused on designing and building a centralised datalake architecture on Google Cloud Platform to serve multiple business domains. The solution successfully integrated data from over 60 heterogeneous sources - including industrial IoT platforms, operational databases, market data APIs, SFTP servers, and cloud storage - with daily ingestion and transformation workflows. The platform was deployed across 6 distinct business units, each with its own 'data as a product' environment, enabling secure, scalable, and cost-efficient access to high-quality data for reporting, forecasting, and operational decision-making. Continuous monitoring ensured ingestion reliability and transformation quality, while maintaining tight cost control on infrastructure and processing.
Designed and maintained scalable ELT data pipelines using Apache Beam/
Dataflow.
Ingested data daily from 60+ sources (REST APIs, DBs, SFTP, S3, Cloud Storage).
Built robust transformation workflows with DBT in BigQuery.
Developed domain-specific 'Data as a Product' platforms.
Managed infrastructure using Terraform.
Implemented CI/CD and monitoring with GitLab pipelines.
Technical Stack: Google BigQuery, Dataflow, Cloud Storage, Pub/Sub, Terraform, DBT, SQL, Python, Apache Beam, Prefect, FastAPI, GitLab, Docker.
Big Query Google cloud Data architecture Data Platform
Myriad-Data
Machine Learning Engineer
ENERGY AND UTILITIES
September 2021 - November 2023 (2 years and 2 months)
Paris, France
The mission focused on developing a smart document processing tool to assist with the treatment of CEE (Certificats d’Économies d’Énergie) submissions. The goal was to automate the extraction, classification, and validation of information from scanned documents submitted by partners and clients, in order to reduce manual review time, ensure regulatory compliance, and improve processing speed and accuracy across large volumes of CEE files.
Developed and maintained deep learning models to:
Classify documents using convolutional neural networks
Detect relevant text zones with YOLOv7 object detection models.
Extract text using PyTesseract OCR.
Designed validation logic to cross-check extracted values and determine
document validity based on consistency and business rules.
Streamlined the operational pipeline to handle large batches of documents
submitted by clients.
Integrated outputs into a structured database for further use in automated and manual validation processes.
Collaborated with cross-functional teams (DevOps, Product, QA) to ensure robust deployment and performance at scale.
Technical Stack: Python, Flask, OCR, Computer vision, Deep learning, TensorFlow, Docker, Git
Machine learning Software Engineering