Data Scientist

Muthoot Finance; Bengaluru; On-site; INR 1,800,000 - 2,400,000; Full time

Job summary

Muthoot Finance in Bengaluru seeks a skilled quantitative modeling specialist to design and deploy end-to-end credit risk models across origination, scoring, PD/LGD/EAD, IFRS 9/CECL frameworks, and stress testing.

You will collaborate with risk, engineering, and data teams to translate model outputs into underwriting strategies, risk pricing, and monitoring, using Python, SQL, PySpark, and cloud platforms.

Qualifications

• Education: Masters or Bachelors in Statistics, Mathematics, Computer Science, or related quantitative field.
• Experience: Minimum 4 years developing statistical and machine learning models in banking, consumer lending, or fintech credit risk.
• Domain Knowledge: Credit lifecycle, delinquency tracking, and credit bureau data structures.

Responsibilities

• Design, build, and recalibrate end-to-end quantitative models across the full credit lifecycle.
• Develop collections and banking tools like Propensity to Buy/Pay, Self-Cure, shadow ratings, and Early Warning Systems.
• Extract and process datasets from credit bureaus and core banking systems; perform WOE/IV analysis and feature selection.
• Validate models with Out-of-Time (OOT) and Out-of-Sample (OOS); monitor drift and performance metrics.
• Ensure governance and documentation per Model Risk Management; defend in reviews and audits.
• Collaborate with MLOps and Data Engineering to deploy models into real-time decision engines and translate outputs into underwriting cut-offs and risk-based pricing.

Skills

Python

SQL

SAS or R

Statistical modeling

Machine learning

Education

Masters/Bachelors in Statistics/Mathematics/CS

Tools

PySpark

Databricks

AWS/Azure/GCP

Job description

Key Responsibilities:

• Credit & Banking Model Development: Design, build, and recalibrate end-to-end quantitative models across the full credit lifecycle, including Application/Behavioral Scorecards, PD/LGD/EAD frameworks, IFRS 9/CECL loss forecasting, and stress testing.
• Advanced Risk & Commercial Strategies: Develop specialized collections and commercial banking tools such as Propensity to Buy/Pay, Self-Cure, shadow ratings, and Early Warning Systems. Integrate key lifecycle models into decision engines for Risk-Based Pricing, dynamic Credit Line Management, and fraud detection scorecards.
• Feature Engineering & Data Pipeline: Extract and process complex structured/unstructured datasets from credit bureaus, core banking systems, and alternative data sources. Perform Weight of Evidence (WOE) transformations, Information Value (IV) analysis, and feature selection.
• Model Validation & Monitoring: Conduct rigorous Out-of-Time (OOT) and Out-of-Sample (OOS) validation. Track model drift, population stability index (PSI), characteristic stability index (CSI), Gini, KS statistics, and AUC-ROC to ensure ongoing accuracy and performance.
• Regulatory Compliance & Governance: Prepare comprehensive model documentation in alignment with Model Risk Management frameworks and defend choices during internal validation and external audits.
• Deployment & Strategy Integration: Partner with MLOps and Data Engineering teams to deploy models into real-time decision engines. Translate model outputs into actionable underwriting cut-offs, credit limit management strategies, and risk-based pricing.

Qualifications & Technical Stack

• Education: Masters or Bachelors degree in Statistics, Mathematics, Computer Science, or a related quantitative field.
• Experience: Minimum 4 years of hands-on experience developing statistical and machine learning models, specifically within banking, consumer lending, or fintech credit risk environments.
• Domain Knowledge: Deep understanding of the end-to-end credit lifecycle (originations, account management, collections), delinquency tracking (Days Past Due - DPD), and credit bureau data structures.

Technical Requirements:

• Programming: Advanced proficiency in Python and SQL. Experience with SAS or R is a plus.
• Data & Cloud Platforms: Hands-on experience working with Big Data tools (PySpark, Databricks) and Cloud platforms (AWS / Azure / GCP).
• Ready tools (PySpark, Databricks) and Cloud platforms (AWS / Azure / GCP).