Senior Data Scientist (Machine Learning & Graph Analytics)

NEBIRU; Brussel Hoofdstad; Sur place; EUR 60 000 - 100 000; Temps partiel

Résumé du poste

NEBIRU in Belgium is seeking an experienced Data Scientist to lead fraud analytics initiatives as a freelancer. You will develop models, analyse large datasets, and collaborate with cross-functional teams to translate business challenges into data-driven solutions.

The role requires 7+ years in data science, strong Python/SQL, Spark, graph analytics, and a production‑oriented mindset. You will design scalable analytics and contribute to evolving fraud analytics practices.

Qualifications

• 7+ years of professional Data Science experience, with significant experience in fraud detection, fraud analytics or a closely related domain.
• Strong hands-on Python and SQL skills, including experience with Pandas and large-scale data processing.
• Proven experience working with Spark or other distributed data-processing technologies.
• Strong knowledge of supervised and unsupervised machine learning, anomaly detection and predictive modelling.
• Practical experience with graph analytics, network analysis or behavioural analytics applied to fraud or risk problems.
• Demonstrated ability to turn complex, high-volume data into robust analytical and ML solutions.
• Experience taking data science solutions towards production, working with software engineering and/or ML engineering teams.
• A proactive, autonomous and business-oriented approach.

Responsabilités

• Develop and continuously improve fraud detection models using machine learning, anomaly detection, graph analytics and behavioural analytics.
• Transform large-scale transactional and customer data into actionable fraud intelligence.
• Lead feature engineering, data enrichment and dataset preparation for fraud analytics.
• Analyse complex fraud patterns, emerging attack vectors and suspicious networks to identify new detection opportunities.
• Design analytical solutions that can operate reliably at scale and contribute to production AI applications.
• Partner with Fraud Operations, Risk, Compliance and Product stakeholders to translate business challenges into effective data-driven solutions.
• Investigate model performance and continuously improve detection accuracy while considering operational impact.
• Contribute to the evolution of fraud analytics practices, methodologies and technical standards.

Connaissances

Python

SQL

Pandas

Spark

Distributed data processing

Machine learning

Graph analytics

Behaviour analytics

Production deployment

Autonomy

Outils

Pandas

Spark

Description du poste

Looking to start a new challenge as a freelancer? Get in touch now.

What you will do

• Develop and continuously improve fraud detection models using machine learning, anomaly detection, graph analytics and behavioural analytics.


• Transform large-scale transactional and customer data into actionable fraud intelligence.


• Lead feature engineering, data enrichment, data quality and dataset preparation for fraud analytics.


• Analyse complex fraud patterns, emerging attack vectors and suspicious networks to identify new detection opportunities.


• Design analytical solutions that can operate reliably at scale and contribute to production AI applications.


• Partner with Fraud Operations, Risk, Compliance and Product stakeholders to translate business challenges into effective data-driven solutions.


• Investigate model performance and continuously improve detection accuracy while considering operational impact and business value.


• Contribute to the evolution of fraud analytics practices, methodologies and technical standards.

What you bring

• 7+ years of professional Data Science experience, with significant experience in fraud detection, fraud analytics or a closely related domain.


• Strong hands‑on Python and SQL skills, including experience with Pandas and large-scale data processing.


• Proven experience working with Spark or other distributed data‑processing technologies.


• Strong knowledge of supervised and unsupervised machine learning, anomaly detection and predictive modelling.


• Practical experience with graph analytics, network analysis or behavioural analytics applied to fraud or risk problems.


• Demonstrated ability to turn complex, high-volume data into robust analytical and ML solutions.


• Experience taking data science solutions towards production, working with software engineering and/or ML engineering teams.


• A proactive, autonomous and business-oriented approach.