Senior MLOps Engineer – Data Platform
Long-term assignment | Mostly remote | 1 week onsite in Brussels every 2 months | Senior

About the opportunity
Within De Cronos Groep, we bring together specialised technology companies and experts to deliver complex digital and technology projects for organisations across a wide range of industries.

For one of our long-term projects, we are currently looking for an experienced Senior MLOps Engineer to take ownership of the design and implementation of a production-grade MLOps capability within an existing enterprise data platform.

This is a long-term, full-time assignment for a senior professional who combines strong MLOps and ML engineering expertise with platform thinking and a genuinely hands-on approach.

The position is predominantly remote, with an expected onsite presence in Brussels for approximately one week every two months. Candidates should therefore be comfortable travelling to Brussels on this recurring basis.

The environment is based on Azure and Databricks and follows a data mesh approach, where domain teams own their data as products while a central platform team provides the governed, reusable and self-service capabilities that enable them to work autonomously.

An MVP data platform is already live with pilot teams onboarded. Machine Learning is the next major capability. Today, a number of models are running outside the central data platform. The objective is to build a production-grade MLOps platform that domain teams can adopt independently for both existing and future ML use cases.

Location: Mostly remote, with approximately 1 week onsite in Brussels every 2 months
Level: Senior
Languages: English required; French is a plus
Duration: Long-term, full-time assignment

The Role
You will take ownership of the MLOps capability — its design, implementation and roadmap — and build it on top of the existing data platform.

You will begin by understanding the current platform, including the existing MLOps setup, landing zones, Unity Catalog, governance and automation. From there, you will assess what can be reused and refine an ML lifecycle that integrates naturally with the platform rather than operating alongside it.

You will then build that capability, working closely with domain teams as your first users.

Databricks and MLflow are at the core of the environment. Deep, hands-on expertise with capabilities such as Unity Catalog model registry, Model Serving, feature engineering, Lakehouse Monitoring and Databricks Asset Bundles is therefore essential rather than simply one skill among many. This is a genuinely hands-on engineering role. You will write the pipelines, templates and tooling yourself and work directly with domain teams to understand where the platform works well and where users encounter friction.

We value demonstrated experience over familiarity with the concepts.

What You'll Do
• Own the MLOps capability end to end — its design, implementation quality and roadmap — and be the person accountable for where it is going.
• Assess the existing platform and refine the MLOps architecture to fit it: determine how models, features and experiments map onto the landing zone, Unity Catalog and governance model already in place, and identify which platform gaps need to be closed before ML can run there.
• Understand and refine the target ML lifecycle design together with the platform architect and domain teams, then deliver it incrementally — first creating a working path for one domain, and subsequently turning it into a paved road that other domains can adopt.
• Implement the full ML lifecycle on Databricks and MLflow — experiment tracking, Unity Catalog model registry, feature tables, packaging, Model Serving and monitoring — tailored for regulated environments.
• Build controlled promotion across development, staging and production using CI/CD technologies such as Azure DevOps / GitHub Actions and Databricks Asset Bundles, ensuring model releases are reproducible and auditable.
• Deliver solutions using off-the-shelf capabilities where they fit and custom components where they do not, and own that engineering judgement.
• Build automated retraining, drift and skew detection using Lakehouse Monitoring or equivalent technologies, together with the alerting and rollback mechanisms required to make these processes trustworthy.
• Productionise both batch and near real-time inference.
• Treat models as data products — with owners, contracts, SLOs and lineage from source data through features to consumers — while feeding health and cost signals into platform-wide observability and governance views.
• Give domain teams cost visibility for ML workloads — including spend attribution per model and domain, right-sized compute, scale-to-zero serving, and visibility into idle endpoints and abandoned experiments.
• Manage ML infrastructure as code using Terraform, following the wider platform standards, and review the ML deliverables produced by domain teams.

What Success Looks Like in the First Year
• Ownership established: you are recognised by the platform team and domain teams as the owner of the MLOps capability and its direction.
• Design agreed: an ML lifecycle architecture that fits the existing platform is reviewed and supported by the platform architect and domain stakeholders within the first quarter.
• First models in production: at least one domain is running monitored, cost-visible models in production through the new paved-road capability.
• Smooth onboarding: a second domain can move a model from experiment to a monitored production endpoint without requiring direct intervention from the central platform team.
• Roadmap delivered: priority MLOps capabilities — for example near real-time inference and automated retraining — are shipped and adopted.

What We're Looking For
• 5+ years in MLOps, ML engineering or platform engineering, with models for which you have built the delivery path and supported in production.
• Someone who can own the design, implementation and roadmap of an MLOps capability through a shared vision — assessing an existing platform, designing to fit it, aligning platform and domain teams behind the direction, and delivering incrementally.
• Deep, hands-on Databricks and MLflow expertise — essential. This includes:
o MLflow tracking
o models and registry in Unity Catalog
o Model Serving
o feature engineering
o Lakehouse Monitoring
o Workflows
o Databricks Asset Bundles
o system tables
You should be able to walk through ML platforms you have personally designed and operated on Databricks.

• Proven delivery of end-to-end ML pipelines using both managed services and custom components, with controlled promotion across environments.
• Hands-on model monitoring that you have built and operated — including drift, skew and performance — rather than simply configured.
• Strong Python engineering skills, producing code that other engineers can maintain.
• Solid Azure and Terraform experience for ML infrastructure.
• Experience with near real-time inference and streaming, including Event Hubs, Kafka and Spark Structured Streaming.
• Working understanding of data mesh and experience building platform capabilities that autonomous domain teams can use.

Nice to have
Experience with any of the following is considered an advantage:
• LLMOps, including evaluation, prompt and version management, RAG, agent frameworks and token- cost management
• AWS
• Kubernetes for model serving
• ML monitoring tools such as Evidently, Arize or WhyLabs
• Experience managing and optimising ML workload costs with quantified results

What You Can Expect
You will work in a platform-as-a-product culture, where successful adoption by domain teams is one of the key measures of success.
You will have real autonomy over what you build and work within an engineering culture that values working code and clear documentation over slide decks.