Senior Engineer, Data Engineering

We are a Brussels-based Data & AI consultancy built around one conviction: the era of the siloed specialist is over. Modern Data & AI projects rarely break because one engineer fails. They break when three excellent specialists ship in isolation and the handoffs collapse. A Solver is a pluridisciplinary Data & AI expert who masters one core domain (Data Foundations, Data Science & AI, or BI & Analytics) and develops a structured understanding of every other discipline in the data flow. When you train a model, you account for the AI Act documentation that will follow. It happens through Cerebro Sessions, our cross-domain knowledge-sharing forum where every Solver participates and contributes. Cerebro is the engine of collaboration at Exsolvæ: how a Data Foundations expert learns what BI & Analytics needs from a model, how an AI engineer absorbs the governance constraints that will land on a project six months later. Cerebro turns a collection of practitioners into a

community, and keeps each Solver broadening their Data & AI knowledge year after year. A community of senior, credible peers across data engineering, AI research, computer vision, and time-series modelling. You work alongside people who challenge your thinking and help you grow. We invest in your ability to deliver with AI, not just talk about it. We expect you to use it, and we help you plan it. You have shipped production pipelines on the Microsoft and Databricks stack. You can defend your architectural choices in front of a sceptical principal engineer. You read papers and release notes when the project warrants it. You understand that a data pipeline exists to serve a business decision, not the other way around. You ask why a metric matters before optimising it. You will operate inside modern data platforms of our clients across Belgium. The mandate is to design, build, and run the data foundation that powers their AI, agentic systems, and decision intelligence programmes. This role is for builders who ship. You will deliver pipelines that ingest, transform, and govern data under regulatory pressure and at scale. Design and build production-grade pipelines on Microsoft Fabric (OneLake, Direct Lake, Mirroring) and Azure Databricks (Lakeflow, Lakebase, Unity Catalog), the two platforms that anchor our Belgian enterprise engagements. ~ Architect medallion-style lakehouses (Bronze, Silver, Gold), applying Data Vault 2.0 for auditability or Kimball where analytical performance is the priority. ~ Implement transformation logic in dbt and PySpark, version-controlled, tested, and documented to a standard that survives external audit. ~ Build the data layer for agentic and GenAI workloads: vector store ingestion, embedding pipelines, RAG-ready datasets, semantic contracts for AI agents. ~ Translate stakeholder questions into data products used in production, not dashboards that get archived. ~ Mentor junior engineers through how you write code reviews. Senior generalists who can pick up new tools quickly are preferred over narrow specialists. If you have shipped production workloads on Microsoft Fabric or Azure Databricks, we want to talk. Cloud and data platform. ADLS Gen2, Azure Data Factory, Synapse Analytics. Languages. Python (advanced, including async, typing, packaging). Azure Data Factory, Databricks Lakeflow, Airflow, Prefect. Data modelling. Data Vault 2.0, Kimball dimensional modelling, Medallion architecture, semantic layer design. Kafka, Azure Event Hubs, Change Data Capture, Real-Time Intelligence in Fabric. AI and agentic data layer. Governance and quality. Great Expectations, Soda, dbt tests. You ship. When a regulator asks for lineage, you show it. When a stakeholder asks where a number came from, you can trace it. You can sit with a non-technical CDO and a principal engineer who reads model architectures, and adjust your register accordingly. Same content, different translation. If you have not benchmarked it, it is not state-of-the-art. If you have not deployed it, it is not production-ready. The pipeline is yours until it is documented and handed off. Junior engineers learn from how you write code reviews, not from a separate mentoring meeting. Languages Fluency in French, English, and Dutch is required, spoken and written. You will run meetings in Dutch, write technical specs in English, and brief francophone stakeholders in French in the same week. A Master's or PhD in Computer Science, Information Management, Applied Mathematics, Physics, or Engineering. ~ Microsoft Azure Data Engineer Associate, Microsoft Fabric Analytics Engineer Associate, or Databricks Certified Data Engineer Professional certification. ~ Public artifacts: papers, patents, open-source contributions, conference talks (FOSDEM, Data Innovation Summit, Devoxx). ~ Track record of delivering complex data platforms at enterprise scale, ideally in regulated environments. ~ Comfort with at least one regulated vertical: financial services, life sciences and pharma, public sector, critical infrastructure. Senior individual contributor freedom. You spend your day on the engagement, not on internal reporting or account management overhead. Responsibility mandates (Lead Solver, Core Domain Leader, Competence Node) recognise different forms of contribution. Time set aside to publish, contribute to open source, or speak at conferences when the work warrants it. The Knowledge & Networking Budget covers it.