A federal judicial police organisation applies data science and artificial intelligence to operational and tactical work supporting investigations into organised crime, cybercrime, terrorism and other serious offences. As a mid-level Data Scientist and ML Engineer, you will build Python machine learning pipelines and bring AI models into production, mainly through on-premise deployments with Azure, Docker and Kubernetes.
The mission
The team develops AI solutions for the wider integrated police service and its national and international partners. The technical landscape covers large structured and unstructured datasets, SQL and NoSQL databases, vector search, deep learning and end-to-end MLOps. Production systems must be scalable, robust, maintainable and aligned with security, ethical and legal requirements.
Your scope runs from system design and model development through deployment, monitoring and maintenance. You will build reusable ML pipelines, version code and models, and work with operational stakeholders to translate investigative needs into reliable AI capabilities. Most deployments are on-premise, while Azure is part of the cloud environment available for suitable workloads.
Your responsibilities
- Design end-to-end machine learning systems that meet scalability, robustness, maintenance and hardware constraints.
- Develop, train and validate models for operational and tactical use, working with large structured and unstructured datasets.
- Build and operate on-premise machine learning pipelines, then move validated models into production with monitoring and maintenance.
- Containerise and orchestrate workloads with Docker, Kubernetes and Kubeflow.
- Establish CI/CD workflows for ML code and models, including versioning, testing and reproducible releases.
- Apply clear documentation, code quality standards and appropriate security, ethical and legal controls.
Your profile
Essential skills
- Bring at least 3 years of professional experience in data science, machine learning and MLOps, including production or large-scale deployments.
- Design and operate on-premise and cloud AI solutions, with hands-on experience in Azure.
- Program confidently in Python and apply PyTorch, TensorFlow and huggingface to machine learning or deep learning work.
- Use Docker, Kubernetes and Kubeflow to containerise and orchestrate ML workloads.
- Work with MySQL, PostgreSQL, Neo4J and Milvus across relational, graph and vector data use cases.
- Implement MLOps and CI/CD workflows with MLflow, Git, GitHub and GitLab, and enforce code quality with ruff.
- Handle large structured and unstructured datasets and communicate technical concepts clearly to non-technical stakeholders.
- Approach problems analytically, collaborate across different profiles and apply careful attention to quality, security, ethics and legal compliance.
Languages
- English: CEFR level not specified, minimum professional proficiency required.
- Dutch or French: CEFR level not specified, minimum professional proficiency in at least one required.
Education
- Master’s degree or doctorate in computer science, artificial intelligence or a related discipline, or equivalent professional expertise.