Senior Data Scientist - AI and Machine Learning

Role Overview: You are looking for a BFSI Machine Learning professional with strong hands-on experience in risk, fraud, underwriting, collections, and cross-sell modeling, along with exposure to GenAI / LLM-based use cases. The role involves end-to-end model development, client interaction, and delivery ownership across large banking and insurance engagements. Key Responsibilities: - Develop and deploy predictive and scorecard models, including: - Cross-sell / up-sell propensity models - Probability of Default (PD), LGD, EAD models - Collections prioritization and recovery models - Insurance underwriting scorecards - Claims fraud detection and leakage analytics - Apply ML/statistical techniques such as Logistic Regression, GBM/ XGBoost, Random Forest, Neural Networks. - Perform feature engineering using transactional, bureau, behavioral, and alternate data. - Conduct model validation and stability monitoring (Gini, KS, PSI, CSI, lift, decile analysis). - Contribute to LLM-enabled BFSI solutions, including: - Automated credit / underwriting note generation - Claims investigation summaries - Risk and audit insight extraction - Hands-on experience with prompt engineering, embeddings, and RAG-based architectures. - Strong hands-on coding in Python (pandas, numpy, sklearn, xgboost, SHAP). - Experience on at least one cloud platform: AWS (SageMaker, S3, Glue), Azure (Azure ML, Azure OpenAI, Data Factory), GCP (Vertex AI). - Exposure to model deployment, monitoring, and MLOps pipelines is a plus. - Work closely with client business, risk, fraud, and underwriting teams. - Translate business requirements into ML problem statements and solution designs. - Support client presentations, model documentation, and insights storytelling. - Mentor junior analysts and ensure delivery quality. Qualifications Required: - 510 years of experience in BFSI analytics / machine learning. - Hands-on experience in credit risk, fraud, underwriting, collections, or cross-sell modeling. - Advanced proficiency in Python. - Working knowledge of LLMs, GenAI, and prompt engineering. - Experience with cloud-based ML platforms. - Strong communication and stakeholder management skills. Additional Company Details: At PwC, you will be part of a vibrant community that leads with trust and creates distinctive outcomes for clients and communities. The purpose-led and values-driven work, powered by technology in an innovative environment, will enable you to make a tangible impact in the real world. PwC rewards contributions, supports wellbeing, and offers inclusive benefits, flexibility programmes, and mentorship to help you thrive in work and life. Together, you will grow, learn, care, collaborate, and create a future of infinite experiences for each other. PwC believes in providing equal employment opportunities and creating an environment where everyone can bring their true selves and contribute to personal and firm growth. Role Overview: You are looking for a BFSI Machine Learning professional with strong hands-on experience in risk, fraud, underwriting, collections, and cross-sell modeling, along with exposure to GenAI / LLM-based use cases. The role involves end-to-end model development, client interaction, and delivery ownership across large banking and insurance engagements. Key Responsibilities: - Develop and deploy predictive and scorecard models, including: - Cross-sell / up-sell propensity models - Probability of Default (PD), LGD, EAD models - Collections prioritization and recovery models - Insurance underwriting scorecards - Claims fraud detection and leakage analytics - Apply ML/statistical techniques such as Logistic Regression, GBM/ XGBoost, Random Forest, Neural Networks. - Perform feature engineering using transactional, bureau, behavioral, and alternate data. - Conduct model validation and stability monitoring (Gini, KS, PSI, CSI, lift, decile analysis). - Contribute to LLM-enabled BFSI solutions, including: - Automated credit / underwriting note generation - Claims investigation summaries - Risk and audit insight extraction - Hands-on experience with prompt engineering, embeddings, and RAG-based architectures. - Strong hands-on coding in Python (pandas, numpy, sklearn, xgboost, SHAP). - Experience on at least one cloud platform: AWS (SageMaker, S3, Glue), Azure (Azure ML, Azure OpenAI, Data Factory), GCP (Vertex AI). - Exposure to model deployment, monitoring, and MLOps pipelines is a plus. - Work closely with client business, risk, fraud, and underwriting teams. - Translate business requirements into ML problem statements and solution designs. - Support client presentations, model documentation, and insights storytelling. - Mentor junior analysts and ensure delivery quality. Qualifications Required: - 510 years of experience in BFSI analytics / machine learning. - Hands-on experience in credit risk, fraud, underwriting, collections, or cross-sell modeling. - Advanced proficiency in Python.