Data Scientist Lead Analyst

As part of Citi’s Financial Crimes and Fraud Prevention - Modeling and Data organization, this role leverages advanced machine learning tools and data mining techniques to identify and combat fraud. A key focus of the role is on data and feature engineering; transforming raw and complex datasets into optimized inputs for developing high-performance fraud models. The role will be responsible for developing and implementing sophisticated fraud models aimed at preventing and mitigating fraud risks across the full fraud lifecycle including application fraud, synthetic ID fraud, account takeover, and evolving fraud attack methods.

The ideal candidate will bring a strong technical background in data processing, feature engineering, and data manipulation, playing a pivotal role in enabling the development of effective and scalable fraud models. The role requires expertise in extracting and engineering key features from large datasets, ensuring that models are not only accurate but also resilient against emerging fraud patterns.

The role will work closely with technology teams, fraud analytics, and various business partners to stay informed about business and technology shifts, identifying both potential and existing fraud impacts. Technical proficiency in model optimization, algorithm development, and real-time analytics is essential for enhancing fraud prevention efforts.

Responsibilities

• Lead data and feature engineering efforts to extract, transform, and prepare high-quality data inputs for fraud model development, focusing on identifying key attributes that drive accurate fraud detection.
• Build predictive models and machine-learning and AI algorithms with large amounts of structured and unstructured data. Ownership and management of fraud models, risk appetite execution and defect analysis.
• Design, develop, and implement advanced machine learning models to detect and prevent fraud across the entire lifecycle, including application fraud, synthetic ID fraud, account takeover, and evolving attack schemes.
• Utilize advanced data processing techniques to manage large, complex datasets, including data cleaning, normalization, and augmentation, ensuring robust model performance.
• Conduct comprehensive exploratory data analysis (EDA) to uncover hidden patterns, trends, and anomalies that can inform model development and feature engineering.
• Collaborate closely with technology teams, fraud analytics, and business partners to align on data strategies, stay updated on industry trends, and proactively identify potential and existing fraud risks.
• Continuously optimize and refine fraud models through feature selection, hyperparameter tuning, and ongoing performance monitoring, ensuring models remain adaptive to new fraud tactics.
• Support model deployment and integration into production systems, ensuring seamless real-time fraud detection and efficient feedback loops for continuous model improvement.
• Evaluate and select appropriate machine learning algorithms and tools based on specific fraud detection needs and data characteristics.
• Engage in cross-functional initiatives to enhance data quality and governance, improving overall fraud prevention capabilities.
• Participate in model validation and testing processes to ensure compliance with regulatory standards and alignment with best practices in fraud risk management.
• Generate and manage regular and ad-hoc reporting to enable effective monitoring and identification of emerging trends.

Qualifications:

• Bachelor’s Degree required in statistics, mathematics, physics, economics, or other analytical or quantitative discipline. Master's Degree or PhD preferred.
• 5+ years in data science, machine learning, or advanced analytics.
• Experience with Generative AI and LLM, preferred
• Strong Technical Skills:

• Proficiency in programming languages such as Python, R, or SQL for data manipulation, feature engineering, and model development.
• Strong experience with data processing tools and libraries (e.g., Pandas, Numpy, PySpark) for handling large and complex datasets.
• Deep understanding of machine learning algorithms (e.g., decision trees, gradient boosting, neural networks, natural language processing) and statistical modeling techniques used for fraud detection
• Expertise in feature engineering, including creating, selecting, and refining features to improve model accuracy and performance.

• Data Engineering: Experience with building and optimizing data pipelines, ETL professes, and real-time data streaming for fraud detection solutions.
• Machine Learning Operations: Familiarity with model development, monitoring, and versioning in production environments.
• Analytics Skills: