T Hub - Data Architect – Microsoft Fabric & Data Intelligence

T-Mobile Polska; Warszawa; On-site; PLN 140,000 - 210,000; Full time

Benefits offered by this job

Medical package

Life insurance

Training platforms

Job summary

T-Mobile Polska seeks a Data Architect to lead the data engine behind the Seamless Network Integration vision. You will design a unified data platform on Microsoft Fabric, harmonizing SIM and network telemetry across our portfolio and partner networks.

You will migrate ~50 TB of legacy Oracle/SAP SIM data into Fabric, build adapters to external CMPs, and deliver secure partner-portal analytics while laying the foundation for real-time streaming and AI capabilities.

Qualifications

• Experience designing enterprise data platforms and data models.
• Proficient in migrating large datasets (Oracle/SAP) to modern cloud architectures.
• Proficient in Python and PySpark for data processing at scale.
• Knowledge of data governance, security, and access controls.
• Experience with real-time streaming concepts is a plus.

Responsibilities

• Design the target Microsoft Fabric state (Lakehouse, Pipelines, Notebooks, Direct Lake).
• Build scalable ingestion pipelines and CMP adapters for high-volume data.
• Implement reporting via Power BI apps and embedded analytics.
• Collaborate with development, QA, and DevOps through CI/CD processes.

Skills

Data architecture

Large-scale data migration

Python

PySpark

ETL/ELT design

Azure Databricks

Azure Synapse

Power BI

Microsoft Fabric

Delta Lake

Tools

Databricks

Synapse

Power BI

Microsoft Fabric

Delta Lake

Job description

We are looking for a Data Architect to lead the data engine behind our Seamless Network Integration (SNI) vision—becoming the global leader in IoT network aggregation. In this role, you will build a unified data platform on Microsoft Fabric by harmonizing SIM and network telemetry across our own portfolio and partner MNO networks. To achieve this, you will migrate ~50 TB of legacy Oracle/SAP SIM data into Fabric, build adaptors into external Connectivity Management Platforms (CMPs), and deliver secure partner-portal analytics while designing the foundation for future real-time streaming (Kafka) and AI capabilities.

Key Responsibilities

• Design the target Microsoft Fabric state (Lakehouse, Pipelines, Notebooks, Direct Lake, Semantic Layer) to migrate ~50 TB of legacy Oracle SIM data and unify Seamless Network Integration data while managing cost, performance, and data residency.
• Build scalable ingestion pipelines (Oracle, CSV, CDC) and CMP adaptors to process high-volume SNI data, structuring it via Medallion Architecture (Bronze/Silver/Gold) and Star Schema models.
• Implement audience-tailored reporting for internal teams and partner portals to deliver real-time data insights using Power BI Apps and Embedded patterns.
• Work with development, QA, and DevOps teams through implementation and CI/CD setup.

Hard Skills

• Deep technical experience with the Azure Data Stack (Databricks, Synapse, Power BI) with a strong focus on or transition to Microsoft Fabric (Lakehouse, Pipelines, Semantic Models).
• Python, PySpark, and large-scale extraction/migration of SIM and network telemetry (Oracle DB >10 TB, batch/CDC, flat files).
• Medallion Architecture, Delta Lake, and Dimensional Modeling (Star Schema) optimized for network and SIM data.
• Entra ID, Keycloak / OAuth2, Azure Key Vault, and Azure networking (Data Gateways, Private Endpoints).
• Granular data security (RLS/OLS), Fabric Git integration, and Azure DevOps CI/CD.

Preferred (Nice-to-Have)

• Hands-on experience with real-time streaming concepts (Apache Kafka, Fabric Eventstream).
• Background in Data Governance (Microsoft Purview), CMP integrations, or legacy SAP BODS replacement.
• Certifications: DP-600, DP-203, or AZ-305.

We offer

• Stable employment based on employment contract.
• Medical package and life insurance.
• Access to training platforms - Percipio, Coursera, Rodos.
• Benefit platform - you choose what you want to use.
• Additional day off for your birthday/nameday.