Research Scientist, Applied Machine Learning - Credit

What you'll do

• Research, design, and develop advanced machine learning methodologies to improve credit risk modeling across underwriting, pricing, credit limits, and portfolio optimization.
• Apply both traditional and cutting-edge ML techniques, including gradient boosting, deep learning, representation learning, sequential modeling, graph learning, reinforcement learning, and large-scale pre-trained models.
• Translate ambiguous credit and business problems into well-defined modeling problems, including hypothesis design, evaluation metrics, and validation frameworks.
• Build and evaluate models using large-scale structured, behavioral, transactional, bureau, repayment, and financial data.
• Develop advanced feature representation and embedding approaches to extract robust signals from heterogeneous data sources.
• Explore reinforcement learning, contextual bandits, sensitivity modeling, uplift modeling, and response modeling for pricing, limit adjustment, and policy optimization.
• Conduct rigorous experimentation, including benchmark comparisons, ablation studies, robustness testing, error analysis, model diagnostics, and interpretability assessments.
• Evaluate models based on predictive performance, stability, generalization, fairness, explainability, operational feasibility, and business impact.
• Assess state-of-the-art research in machine learning, deep learning, reinforcement learning, causal inference, and financial AI, and adapt relevant methods to credit risk use cases.
• Partner with Risk, Data Science, Product, and Engineering teams to translate validated research prototypes into scalable, production-ready solutions.
• Contribute to internal modeling standards, research documentation, reproducibility practices, and knowledge sharing.

Requirements

• PhD in Computer Science, Machine Learning, Artificial Intelligence, Quantitative Finance, Financial Engineering, Computational Finance, or a closely related discipline at a reputable university with strong academic credentials.
• Prior experience (e.g., full-time/part-time/internship) working on applied machine learning research or model development, particularly in areas such as credit risk modeling, financial AI, deep learning, reinforcement learning, causal inference, graph learning, sequential modeling, or large-scale predictive modeling.
• Prior experience  (e.g., full-time/part-time/internship) working in financial, fintech or internet / technology industry is a plus.
• Strong experimental rigor, including hypothesis formulation, benchmark design, ablation studies, error analysis, robustness testing, model diagnostics, and interpretability analysis.
• Ability to read, evaluate, reproduce, and adapt state-of-the-art research papers for practical modeling applications.
• Ability to explain complex modeling trade-offs to technical and non-technical stakeholders across Risk, Data Science, Product, and Engineering.