Data Scientist II

Role Overview 

The Data Scientist II will be responsible for developing, validating, and optimizing internal fraud and authentication models that support TD's objective of expanding proprietary risk capabilities and reducing reliance on external vendor models. The role covers the end-to-end model development lifecycle, including data discovery, feature engineering, experimentation, model evaluation, documentation, business impact assessment, and productionization support. 

This is a hands-on development role with significant responsibility for presenting results and communicating with technical and non-technical stakeholders. The successful candidate will partner with Strategy, Engineering, Product, Data, Model Validation, and Risk teams to translate business challenges into scalable analytical solutions. 

Key Responsibilities 

• Develop, validate, and optimize machine learning and statistical models for fraud detection, authentication, identity verification, account opening, account takeover, scam detection, and related risk use cases. 

• Lead end-to-end model development activities, including data discovery, exploratory analysis, data preparation, feature engineering, experimentation, model training, hyperparameter tuning, performance evaluation, and business impact assessment. 

• Evaluate internal models against vendor models and existing business strategies, using appropriate fraud and risk metrics such as capture rate, false-positive rate, lift, precision, recall, stability, and customer friction. 

• Build reproducible and well-documented analytical workflows using Python, SQL, and enterprise data platforms, and support implementation and productionization with Engineering, FICO, MLOps, and platform teams. 

• Perform model monitoring, stability and segmentation analysis, performance diagnostics, and issue investigation to identify degradation, data gaps, and remediation actions. 

• Prepare model documentation and analytical evidence for Model Validation, Model Risk Management, governance reviews, approvals, and lifecycle management. 

• Analyze customer, authentication, and fraud behaviour to identify emerging trends, risk gaps, feature opportunities, and areas where internal models can improve detection coverage. 

• Prepare and deliver executive-ready presentations, technical documentation, reports, and business impact summaries for technical and non-technical audiences. 

• Lead cross-functional discussions to align stakeholders on data requirements, model scope, success criteria, risks, dependencies, timelines, implementation considerations, and next steps. 

• Contribute to continuous improvement of internal modeling standards, reusable analytical frameworks, development practices, and proprietary fraud and authentication capabilities. 

Qualifications  Required 

• Undergraduate or graduate degree in Data Science, Statistics, Mathematics, Computer Science, Engineering, or a related quantitative field. 

• 3-5+ years of experience in data science, machine learning, statistical modeling, or advanced analytics. 

• Strong hands-on proficiency in Python, SQL, machine learning, statistical analysis, feature engineering, and data visualization. 

• Experience developing and evaluating predictive models in a business or production-oriented environment. 

• Strong presentation, written communication, and stakeholder management skills. 

• Ability to explain technical methods, model results, limitations, and business impact to different audiences. 

• Ability to independently manage development activities across multiple initiatives in a fast-paced environment. 

Preferred / Strong Assets 

• Previous experience with Model Validation (MV), model risk, model governance, or independent model review processes. 

• Experience in fraud analytics, authentication, cybersecurity, financial crime, identity verification, or digital banking. 

• Experience developing fraud, risk, or customer behaviour models using highly imbalanced datasets. 

• Experience with Azure Databricks, Azure Machine Learning, MLOps pipelines, Streamlit, FICO platforms, or other cloud-based analytics environments. 

• Experience supporting model implementation, production deployment, performance monitoring, and lifecycle management. 

• Familiarity with graph analytics, network analysis, or advanced AI techniques is an asset. 

•  modeling capabilities.