Data Engineer

Markit is how procurement teams across Europe buy IT products. Around 230 people, 50+ countries, one platform. Every order, price, supplier and customer runs through it, and we want that data to work for the whole company.

Our data platform starts with you: a warehouse on Azure, pipelines from the marketplace's service databases (mostly PostgreSQL), our ERP and the CRM, and an access model that lets people and AI tools use the data safely. The stack is open. We expect you to propose it and defend the choice. You sit in the platform team, which runs our Azure infrastructure with you.

The warehouse also opens a product. Enterprise customers want to understand their IT spend, and better reporting on the data we already hold is an opportunity we are looking at. You help shape what that becomes.

Responsibilities

• Design and build the data warehouse on Azure, with a stack you propose and a written decision behind it.
• Build and run pipelines from the marketplace, the ERP and the CRM, with monitoring, data quality checks and documentation. Agree data contracts with the service teams so analytics does not depend on their internal tables.
• Move internal reporting onto the warehouse: build the data models that dashboard tools such as Power BI run on, and the first reports until we have an analyst.
• Build safe access for AI tools and agents: read-only, scoped, logged.
• Work with product and sales on customer-facing spend reporting, and introduce the warehouse internally so people know where the data is and how to use it.

The challenge

• You build from the ground up. The list of reports people want is long, so you pick the order and say what waits.
• Source systems are in the middle of a platform migration. Schemas change under you.
• Not every source has a good API. Some data you will have to pull the hard way.
• Customer and personal data from 50+ countries. GDPR and access control are part of the design from day one.
• You are the first data engineer. What you build has to be understandable by the people who come after you.

Requirements

• 5+ years in data engineering, including at least one warehouse or lakehouse you built or rebuilt and ran in production.
• Strong SQL and Python, data modelling (dimensional or similar), and hands-on experience with orchestration and ELT tooling. You treat pipelines as code: version control, tests, CI/CD.
• Experience with Azure data services, or a clear case for what you would run on Azure instead. You have pulled data out of operational PostgreSQL databases without hurting them (CDC, logical replication or similar).
• You have handled personal data under GDPR: access control, retention, masking.
• You explain data to non-technical people and write things down.