Data Scientist – FEC Transaction Monitoring
Preventing Money Laundering & Terrorist Financing are the key aspects for society and the organization, and as a bank, the team plays a major role in this field. In order to be effective in fighting financial crime, the team needs smart individuals like yourself who get their energy from looking into data, patterns, and behavior of the customers in order to uncover illegal activities.
You will be responsible for optimizing the current Transaction Monitoring (TM) system and ensuring it is future-proof for introducing more advanced analytics methodologies such as Machine Learning into the TM models.
The TM system evaluates millions of transactions every single day and needs to perform within the boundaries that have been set with respect to computing time and resources. The model is in ongoing development and new (computationally intensive) functionality is being added. Your responsibility would be to optimize the performance of model components. The challenge is that this is not solely a technical assignment. Working in the context of transaction monitoring requires you to thoroughly understand the functionality of the model and its components.
The candidate will collaborate with the team to help contribute to delivering the following results:
- Build functionality into the data model (TDM) while continuing to combine all (30+) data sources required for Transaction Monitoring in a computationally efficient way. The sources contain event data [e.g., transaction data and logging data from sources like CNA, AZS, PEX, and RASS] and enrichment data (e.g., customer data, and crash data from sources like Siebel, Omnikassa, Geldmaat, VCM)]
- Ensure model component design enables the bank to comply with all regulatory, audit, and other internal requirements;
- Set up the model components to function as solid building blocks that can function on their own, and can be leveraged by other (KYC/TM) projects as well;
- Ensure the computational performance of the Transaction Monitoring solution is sufficient for daily batch scoring and for one-off runs on multiple years of data (using (Py) Spark).
The ideal candidate:
- Is motivated to work on models that realize measurable impact (i.e. catch criminals), not necessarily to build the most complex advanced analytics models;
- Is structured, precise, communicative, and can work well with different people and roles;
- Thrives in a high-impact, dynamic and high-paced environment;
- Is pro-active, has a business focus, and can-do mentality to make a difference in fighting financial crime.
For this position, it’s required:
- Academic degree (MSc / Ph.D.) in Data Science, Computer Science, Mathematics, or a related field (required);
- Software development skills such as Git, bash scripting, and release management and proven success in bringing analytics models to production (required);
- A minimum of 5 years of experience in advanced analytics/modeling / artificial intelligence activities working with large data sets (required);
- Excellent verbal and written communication skills in English (required);
- Highly skilled in Python, PySpark, and Spark (required).
Nice to have
- Skilled in working within Azure Databricks (preferred).
Then we would like to receive a motivation showing the required competencies together with your updated CV. Do you have questions regarding this role? please reach out to Angie Hollink via email@example.com