Lead Data Engineer at Allianz · open to new roles

MohamedHakkache

I build the systems that move data from wherever it's trapped to wherever decisions get made. Right now that means lifting a decades-old mainframe and SAS insurance portfolio onto a Databricks lakehouse — with lineage, quality gates and CI/CD that hold up in production.

Explore projects Get in touch Resume ↓

pipeline · career_run

SOURCEmainframe · sas · erp

BRONZEraw ingest

SILVERcleaned · conformed

GOLDbusiness-ready

SERVEdashboards · models

lineage: unity catalog ● all checks passed

Years in data

Certifications

Clouds in production

Countries worked with

Technologies in daily use

About

Engineering first, then data

I trained as an industrial engineer, which is why I think about pipelines the way I was taught to think about production lines: throughput, failure modes, and the cost of a defect found late.

Eight years in, I've gone from analyzing production data on a factory floor to leading the migration of insurance portfolio processing off a mainframe. The through-line is the same: find where information gets stuck, and unstick it permanently rather than with a report.

Lately much of my attention goes to AI inside the pipeline — not chatbots bolted on top, but LLM-assisted enrichment, generated test scenarios, and synthetic data for validating pipelines before real data exists. I build in the open; OpenLakehouse is where most of it ends up.

Stack

What I work with

Grouped the way a pipeline actually runs rather than by vendor — because the interesting question is never "do you know Azure", it's "where does the data stop moving".

IngestLanding raw data

Lakeflow ConnectAutoloader Azure Data FactoryCloud Data Fusion DataflowKafkaDebezium CDCREST APIs

StoreTables and formats

Delta LakeApache Iceberg Azure Data LakeOneLakeBigQuery SnowflakeAzure SQLSQL Server OracleMongoDBMinIO

ProcessRaw into usable

PySparkDatabricksDLT / Lakeflow Spark SQLdbtPython SQLScalaTrino Azure SynapseMicrosoft Fabric

OrchestrateScheduling and shipping

Apache AirflowDagster Databricks Asset BundlesCloud Composer Azure DevOpsGitLab CIDocker Kubernetespytest

ServeWhere people meet data

Power BILooker StudioTableau MicroStrategySuperset Semantic modelsKPI design

Govern & observeTrust and uptime

Unity CatalogMicrosoft Purview Entra ID / RBACData quality frameworks LineagePrometheusGrafana ElasticsearchKibana

AI in the pipelineNewer ground

RAGVector searchEmbeddings LLM data enrichmentAI-generated tests Synthetic test dataClaude CodeGitHub Copilot

Experience

Career pipeline

Six roles, one direction: closer to the data, and closer to the decisions it feeds.

Lead Data Engineer Now

Sep 2025 — Present

Allianz · Freelance

Mainframe & SAS → Azure Databricks migration

Migrated legacy mainframe/SAS insurance portfolio processing onto a modern Azure Databricks lakehouse.
Designed a multi-layer architecture with full traceability from source system to business-ready tables.
Built a data quality framework with automated validation, anomaly detection and quarantine handling.
Orchestrated end-to-end pipelines with PySpark notebooks, DLT and Databricks Lakeflow.
Set up automated testing, continuous integration and infrastructure-as-code deployment across dev and prod.
Governed lineage and access through Unity Catalog; monitored runs, investigated incidents, added alerting.
Led data engineers on pipeline design, development standards and the platform roadmap.

Azure DatabricksDelta LakeDLT Unity CatalogPySparkdbt pytestJira / Zephyr

Data Engineer

Apr 2025 — Sep 2025

Expleo Group · client: Renault Group

Romania · France · Spain — Firmware Over-The-Air testing

Designed and automated pipelines in Google Cloud Data Fusion and Dataflow for large-scale vehicle test and anomaly datasets.
Built BigQuery transformations tracking campaign performance, defect rates and system reliability.
Delivered Looker Studio dashboards for test coverage, campaign efficiency and anomaly resolution.
Coordinated across three countries in Agile ceremonies, aligning data workflows with testing operations.

BigQueryData FusionDataflow Looker StudioAirflowPythonGitLab

Data Engineer

Sep 2023 — Apr 2025

Emirates Logistics

Tangier, Morocco

Architected the company's business intelligence ecosystem on Microsoft Fabric and Power BI from scratch.
Developed pipelines in Fabric Data Factory and Azure Data Factory, centralizing multiple business systems.
Optimized transformation notebooks with PySpark across Databricks and Fabric Lakehouse.
Managed roles, security and governance through Microsoft Entra ID and Purview.
Trained end-users, driving self-service BI adoption across departments.

Microsoft FabricPower BIAzure Data Factory PySparkAzure SQLPurview

Data Engineer

Mar 2021 — Apr 2023

APM Terminals MedPort Tangier

Tangier, Morocco

Led the migration from on-premise BI to Azure — Synapse Analytics, Data Factory, Power BI Service.
Built ETL pipelines integrating operational systems into Azure SQL and Data Lake Storage.
Delivered standardized dashboards for global terminals in Los Angeles, Algeciras and Rotterdam.
Established BI governance and data quality frameworks, reducing inconsistencies across business reporting.

Azure SynapseAzure Data FactoryPower BI ScalaPySparkMicroStrategy

Data Analyst

Jan 2020 — Mar 2021

APM Terminals MedPort Tangier

Tangier, Morocco

Analyzed large operational datasets with Python and SQL for operations, maintenance and supply chain teams.
Delivered automated MicroStrategy dashboards covering traffic flow, safety and productivity.
Ran Lean/Kaizen process analysis to identify efficiency gaps and drive cost control.

PythonSQLMicroStrategy Excel / VBAStatistics

Product Engineer / Data Analyst

Sep 2018 — Jan 2020

LEAR E-Systems

Tangier, Morocco

Analyzed production data to evaluate material usage, process times and cost structures.
Maintained technical documentation — BOMs, routing, APD times — with cross-system accuracy.
Reported monthly project KPIs on cost, quality and timing to management.

SQLPythonPower BI BOM analysisPFMEASix Sigma

Projects

Built in the open

Side projects are where I test ideas that would be irresponsible to try first on someone's production data.

Flagship · 39 services · one docker compose up

OpenLakehouse

A self-hosted, open-source data platform that does what a cloud lakehouse does — Spark writing real Iceberg tables, Trino querying them, Kafka and Debezium streaming change data, Dagster orchestrating, Superset visualizing, MLflow tracking models — all behind one sign-on, running entirely on your own hardware for nothing.

See how it's built