Sr Data Scientist – Financial Crimes & Fraud Prevention
| Company | Citigroup |
|---|---|
| Location | O’Fallon, MO, USA, Tampa, FL, USA, San Antonio, TX, USA, Florence, KY, USA, Johnson City, TN, USA, Jacksonville, FL, USA, Wilmington, DE, USA |
| Salary | $113840 – $170760 |
| Type | Full-Time |
| Degrees | Bachelor’s |
| Experience Level | Senior |
Requirements
- Bachelor’s degree in engineering, Statistics, Economics, Finance, Mathematics or a related quantitative field from a premier institute required.
- Minimum 5+ relevant experience in data analysis, data mining, or statistical analysis.
- Must have a working knowledge of Python, SQL, Teradata, RDBMS, Hadoop/Hive Tools.
- Experience in statistical analysis with working knowledge of at least one of the following statistical software packages: Python (Preferred), SQL, SAS (required)
- Experience with AI/ML Frameworks (eg TensorFlow, PyTorch , Scikit-learn)
- Knowledge of Large language models (LLM’s) for text analysis and Fraud Intelligence
- Experience in Predictive modeling, statistical analysis and machine learning techniques.
- Ability to analyze large-scale, unstructured data and generate actionable insights
- Familiarity with digital fraud detection tools and experience working with cybersecurity datasets and threat intelligence platforms.
- Prior experience in developing dynamic dashboards using visualization tools such as Tableau.
- Data Science work in any risk domain will be preferable.
- Experience in identifying fraud patterns in large consumer banking portfolios.
- Successful candidate will have a demonstrable analytic, problem solving, and leadership skills, and has the ability to deliver projects in a fast-paced environment.
- Excellent quantitative and analytic skills and data-driven mindset; ability to derive patterns, trends and insights, and perform risk/reward trade-off;
- Ability to effectively collaborate with cross-functional partners and management
- Solutions-oriented “can do” attitude, with ability to drive innovation via thought leadership while maintaining end-to-end view.
- Extremely detail-oriented, with strong, intellectual curiosity. Ability to effectively multi-task and work in a fast-paced and evolving environment, while setting meeting high standards.
Responsibilities
- Creating analytical solutions to mitigate losses across all LOB’s utilizing various statistical/advanced data science techniques.
- Analysis of compromised cards data from dark web to identify emerging fraud trends, detect suspicious fraud patterns, and anomalies, identify high risk merchants, locations and transaction behaviors.
- Leverage AI/ML models (eg anomaly detection, graph neural networks, NLP) to anticipate fraudulent behavior.
- Experience with AI techniques and implement AI powered automation to improve fraud detection efficiency.
- Develop and enhance data models and algorithms to identify high risk accounts for proactive monitoring and closure. Additionally, assess the impact of compromised cards on fraud losses, use statistical analysis to quantify risk exposure and identify abnormal spending patterns.
- Collaborate with threat intelligence teams to incorporate external fraud signals into risk models. Identify fraud rings, mule accounts, synthetic identities by linking compromised data to existing customer portfolios.
- Generate executive level insights and reports for leadership team using advanced visualization techniques and provide regular updates on fraud trends and emerging threats. Provide actionable insights to senior global stakeholders by leveraging data analytics and reporting.
- Lead POC’s with new vendors, evaluating fraud detection tools, data enrichment platforms and dark web monitoring solutions.
- Perform ad-hoc analysis on large, unstructured datasets (eg transaction logs, dark web feeds) to identify fraud indicators. Use Python, SQL and SAS to extract, transform and analyze complex datasets.
- Manage significant fraud events by helping coordinate information sharing across financial crime and fraud prevention Org. Partner with various cross-functional teams such as Fraud Policy, Analytics & Modelling, Security Operations Center to help design intelligence derived solutions to detect fraud.
- Collaborate with fraud analytics modelling function to understand new fraud detection capabilities, develop new analytical solutions leveraging unstructured data sets and variables.
Preferred Qualifications
- Master’s degree not required but beneficial.
- Data Science work in any risk domain will be preferable.