Senior Data Scientist

Job Description

As a Data Scientist focusing on credit risk modelling and financial risk analysis, your primary role will involve building and enhancing models for various aspects of credit risk management. You will work on developing models for credit risk assessment, underwriting, customer eligibility, risk-based pricing, credit limit determination, delinquency prediction, and portfolio monitoring. Additionally, you will be responsible for creating models related to Probability of Default, bureau score enhancement, early warning signals, collections prioritization, fraud risk assessment, loss forecasting, and customer affordability evaluation.Key Responsibilities:- Develop models incorporating bureau data from sources like CIBIL, Experian, Equifax, CRIF, or similar credit bureaus.- Engineer bureau-based features including credit utilization, enquiry patterns, repayment behavior, credit vintage, open and closed tradelines, delinquency history, unsecured exposure, EMI burden, settlements, write-offs, and defaults.- Support financial risk modelling activities such as vintage analysis, cohort analysis, roll-rate analysis, expected loss estimation, portfolio risk assessment, and stress testing.Qualifications Required:- 6-7 years of experience in data science, with a preference for backgrounds in fintech lending, digital lending, NBFC, banking, credit analytics, AdTech, growth analytics, performance marketing, or consumer internet.- Strong hands-on experience in bureau data analysis, credit risk modelling, propensity modelling, affinity modelling, customer segmentation, campaign modelling, and ML-based decisioning.- Good understanding of lending concepts like underwriting, eligibility criteria, approval processes, disbursal mechanisms, delinquency management, collections strategies, portfolio risk evaluation, customer affordability assessment, and repayment behavior analysis.- Proficiency in working with alternate data sources such as behavioral data, clickstream data, app/web events, bank statements, transaction data, campaign data, CRM data, device signals, and third-party data.- Sound knowledge of ML and DL algorithms encompassing classification, regression, clustering, recommendation systems, ranking, uplift modelling, time-series forecasting, and anomaly detection.- Strong programming skills in Python and SQL, along with experience in libraries like pandas, NumPy, scikit-learn, XGBoost, LightGBM, CatBoost, TensorFlow, or PyTorch.- Familiarity with key business metrics including approval rate, conversion rate, default rate, delinquency rate, repayment rate, customer acquisition cost (CAC), cost per acquisition (CPA), return on advertising spend (ROAS), customer lifetime value (LTV), click-through rate (CTR), conversion rate (CVR), and risk-adjusted profitability assessment.- Proficient in translating complex business challenges into analytical frameworks and scalable modelling solutions.- Strong stakeholder management and communication skills to effectively collaborate with various teams and convey analytical insights.This job opportunity provides a challenging and rewarding environment for a seasoned Data Scientist to leverage their expertise in credit risk modelling and financial risk analysis within a dynamic industry landscape. As a Data Scientist focusing on credit risk modelling and financial risk analysis, your primary role will involve building and enhancing models for various aspects of credit risk management. You will work on developing models for credit risk assessment, underwriting, customer eligibility, risk-based pricing, credit limit determination, delinquency prediction, and portfolio monitoring. Additionally, you will be responsible for creating models related to Probability of Default, bureau score enhancement, early warning signals, collections prioritization, fraud risk assessment, loss forecasting, and customer affordability evaluation.Key Responsibilities:- Develop models incorporating bureau data from sources like CIBIL, Experian, Equifax, CRIF, or similar credit bureaus.- Engineer bureau-based features including credit utilization, enquiry patterns, repayment behavior, credit vintage, open and closed tradelines, delinquency history, unsecured exposure, EMI burden, settlements, write-offs, and defaults.- Support financial risk modelling activities such as vintage analysis, cohort analysis, roll-rate analysis, expected loss estimation, portfolio risk assessment, and stress testing.Qualifications Required:- 6-7 years of experience in data science, with a preference for backgrounds in fintech lending, digital lending, NBFC, banking, credit analytics, AdTech, growth analytics, performance marketing, or consumer internet.- Strong hands-on experience in bureau data analysis, credit risk modelling, propensity modelling, affinity modelling, customer segmentation, campaign modelling, and ML-based decisioning.- Good understanding of lending concepts like underwriting, eligibility cr

Keyskills :
data sciencebankingconsumer internetPythonSQLNumPyclassificationregressionclusteringanomaly detectionstakeholder managementcommunication skillsfintech lendingdigital lendingNBFCcredit analyticsAdTechgrowth analyticsperformance marketing