Data Scientist - Director - Data & Analytics Engineering

Morgan Stanley is seeking a Data Scientist to join the CDRR Technology team within the Fraud Department. The position is responsible for the development of statistical and machine learning models used to identify, assess, and mitigate fraud risk across Morgan Stanley products.   The successful candidate will independently execute model development assignments, including exploratory data analysis, analytical dataset construction, feature engineering, algorithm selection, model training, performance evaluation, and technical documentation. The role requires demonstrated expertise in the mathematical and statistical foundations of classical supervised and unsupervised machine learning methods and the ability to apply those methods to large, complex, and imperfect real-world datasets.   This is a Director-level individual contributor position and does not include formal people-management responsibilities. The individual will manage assigned model development projects with guidance from senior members of the team. The individual will also provide technical guidance and mentoring to junior Data Scientists and Data Engineers.   Since 1935, Morgan Stanley is known as a global leader in financial services, always evolving and innovating to better serve our clients and our communities in more than 40 countries around the world.   What you'll do in the role: ●        Independently execute end-to-end model development, including technical documentation. ●        Develop and evaluate models for highly imbalanced, non-stationary, and adversarial environments where fraud patterns, customer behaviour, and operational processes evolve over time. ●        Apply statistical and mathematical principles to model selection, validation, and performance assessment, including the treatment of class imbalance, overfitting, data leakage, missing data, and model stability. ●        Monitor deployed models, diagnose performance degradation, assess emerging fraud patterns, and recommend recalibration, redevelopment, threshold changes, or retirement as appropriate.   What you'll bring to the role: ●        6+ years of professional experience in data science, machine learning, statistical modeling, or quantitative analytics. ●        Demonstrated depth of knowledge in statistical inference, probability, sampling, hypothesis testing, regularization, bias-variance trade-offs, optimization, feature selection, dimensionality reduction, model calibration, and statistical diagnostics. ●        Hands-on experience performing exploratory data analysis, constructing analytical datasets, and engineering features from large, complex, and imperfect real-world data. ●        Advanced proficiency in Python and SQL, and experience with Hadoop, Hive, Impala, Spark, or PySpark. ●        Experience designing statistically valid training, validation, and testing approaches and evaluating models using appropriate performance, calibration, stability, and diagnostic measures. ●        Demonstrated ability to independently manage model development assignments and produce technical documentation suitable for review and governance. ●        Excellent written and verbal communication skills. Including the ability to communicate complex analytical methods and results to technical and non-technical stakeholders.   Good to have: ●        Experience developing machine learning or statistical models for fraud detection, financial crime, transaction monitoring, payment risk, or anomalous-behavior detection. ●        Experience analyzing high-volume transactional, account, client, or behavioral data within financial services. ●        Knowledge of fraud typologies, risk indicators, and security issues applicable to banking or Wealth Management. ●        Experience developing models within a regulated environment subject to Model Risk Management,