One of our blue chip clients based in Brussels require a Data Scientist to join them initially till the end of 2024 with a view to a 12-month extension thereafter. See below for further information
Job Description
Develop and implement data science use cases, leveraging supervised and unsupervised machine learning techniques. Collaborate closely with cross-functional teams to identify business opportunities related to financial crime. Identify the data that are required, collect them, understand them and do the data preparation work (cleansing, enrichment, …) so they can be processed Implement predictive models to score and prioritize the alerts raised by the existing and future anomaly detection systems Develop anomaly detection algorithms using supervised and unsupervised machine learning models or any other relevant data analytics techniques such as Natural Language Processing or graph/network analysis Analyse how the actors of our ecosystem of counterparts (clients, agents, beneficial owners, …) interact using graph databases and network analytics techniques, and detect unsuspected and potentially risky relationships Iteratively improve the performance of the models in close collaboration with subject matter experts Design both static and interactive data visualizations to communicate, in a clear and impactful manner, the results of your work to stakeholders who may not necessarily have a strong background in data science
Tasks:
Re-calibration of NetReveal scenario parameters Perform quantitative analysis on improvements in our AML scenarios from NetReveal including Above-The-Line testing, avoiding overlapping alerts etc. Improving investigation tooling Working out strategy/roadmap to improve current investigation tooling for AML NetReveal alerts Develop “insight” dashboard Develop an insight dashboard to follow up on financial crime, market abuse, sanction circumvention etc. trends and KPIs from our core activities. TM Beyond: decreasing false positive alerts from NetReveal (Use Case 3) Supervised machine learning solution to decrease number of false positive alerts generated by NetReveal transaction monitoring system Automate closing of false positive hits from Norkom PoC on a solution to automatically close large volumes of false positives hits generated by our sanction screening process. In addition, this project includes other data initiatives to improve overall efficiency of the hit generation process.
Technical skills:
Master’s or doctoral degree in a quantitative field (, computer science, artificial intelligence, mathematics, physics, statistics). Minimum of 2 years of experience in developing machine learning models Strong programming skills in Python, SQL, and Spark. Expertise with Scikit-learn and Pandas libraries. Proficiency in data science techniques and know-how to use them in a business context
Bonus Points:
Experience in building data products from design to production. Any experience you have with any of the following will be a benefit: Power BI Hadoop ecosystem Git Proficiency in visualization libraries/tools (, Plotly/Dash/NetworkX). Familiarity with network analytics. Knowledge of compliance-related domain (sanctions, transaction monitoring, KYC, fraud detection, …) ideally in the financial industry. Previous exposure to generative AI and/or Natural Language Processing. Familiarity with cloud platforms, preferably Azure
