Mouhammad
Diakhate

Analyze and improve the data platform for European customers across France, Hungary, Germany and the Netherlands.

• Collaborate closely with customers to migrate existing data workloads to Databricks
• Leading migration strategy for 300+ batch legacy workloads to Databricks streaming pipelines
• Maintain legacy workloads and make existing Data Platform IaC more user-friendly

Proof of concept: challenged against another team on how to build Elia's settlement application — led the rebuild from scratch with Spark to prove feasibility while ensuring reliability.

• Worked closely with tech and business stakeholders to validate accuracy, scalability and performance
• Designed and built the Spark-based settlement engine from scratch
• Ran it all on a low-budget on-prem setup — a single 4-core / 16GB node
• Delivered 25 settlement products (5 Levies + 20 ADREA) with complex business rules
• Stress-tested on 4M+ records: under 15-minute total execution time
• Reached 78% functional and architectural coverage within PoC scope

Led Sogelink's architecture improvement while helping Groupe SEB maintain and evolve their data platform.

Sogelink

• Advised stakeholders on migration strategy from a legacy Data Lake to a modern Lakehouse
• Led the migration of the Data Lake to Apache Iceberg
• Integrated new CRM/ERP data sources from Dutch and Norwegian subsidiaries
• Refactored Terraform IaC to streamline developer experience

Groupe SEB

• Developed API ingestions for new online and retail marketplaces across six countries
• Maintained and enhanced Retail pipelines — monitoring, incident management, stakeholder alerts

Set the direction for redesigning legacy EDI preprocessing into a robust data processing tool for container stowage optimization.

• Built an EDI parser and data model, applying SOLID design patterns
• Refactored legacy business rules, cutting code volume by 66%
• Designed a "restow" algorithm to optimize heavy-container placement
• Prepared a migration strategy and usage guidelines to transfer knowledge internally

Migrated legacy BI from SQL Server (AWS RDS) to a Databricks Lakehouse architecture.

• Designed a hybrid data vault and medallion Lakehouse architecture on Databricks
• Built a 3-hop architecture with an SCD2 reliability layer and Change Data Feed
• Migrated legacy SQL Server workloads to production-ready gold tables for Power BI and SSRS
• Delivered a data lineage layer and full functional documentation for all migrated jobs
• Optimized production performance through Spark UI monitoring and tuning

Advised on new KPI development for the Transilien T-REX project.

• Designed cost-effective Azure cloud architecture and secured cloud resources
• Managed Databricks and Data Factory pipeline scheduling, execution and monitoring
• Built a star-schema Data Warehouse on Azure, orchestrated via Data Factory
• Migrated the CD pipeline artifact repository from Nexus to JFrog Artifactory

Automated Cloud infrastructure provisioning and observability of application logs.

• Automated Cloud infrastructure provisioning with Terraform, Bash and Nginx scripts
• Built a serverless Kafka pipeline for log ingestion, with source/sink connectors
• Delivered a self-service gateway and template registry for infra provisioning requests