Data Scientist

Agam Capital India; Mumbai; On-site; INR 1,200,000 - 2,500,000; Full time

Job summary

Agam Capital India is seeking a Data Scientist to join its AI team focused on consumer lending and insurance analytics in Mumbai. You will build and refine statistical and machine learning models, connect outputs to financial results, and integrate them into agentic AI workflows.

The role covers product improvement, renewals, cross-sell, model validation and loan risk, with emphasis on data engineering, feature creation and robust evaluation.

Qualifications

• Bachelor's or master's in a quantitative field.
• Strong Python and SQL with pandas, NumPy, scikit-learn.
• Experience with cross-validation and data leakage prevention.
• Experience developing LLM agents with adaptive planning.

Responsibilities

• Develop and improve statistical and machine learning models for product performance and credit risk.
• Explore and link customer, loan, policy and campaign data; engineer features.
• Evaluate models against baselines; assess calibration and performance.
• Combine model outputs with financial projections for strategy assessment.
• Document assumptions and collaborate with engineering on deployment and monitoring.

Skills

Python

SQL

Pandas

NumPy

scikit-learn

XGBoost/LightGBM

Feature Engineering

Cross-validation

Education

Bachelor's or Master's in statistics / math / CS / engineering

Tools

pandas

NumPy

scikit-learn

XGBoost

LightGBM

Job description

Mandatory to have experience with bank / NBFC / lending / credit risk

About Us

Agam develops financial and actuarial analytics through pALM, its proprietary asset and liability management platform. Our work combines quantitative modeling, financial projections and AI to support decisions on products, customers and risk.

Role Overview

We are looking for a Data Scientist to work as part of our AI team on consumer lending and insurance analytics. You will develop and improve statistical and machine learning models, connect their outputs to financial outcomes, and integrate them into agentic AI workflows. The work covers product improvement, experience analysis, renewals, cross-sell, model validation and loan risk.

Key Responsibilities

• Develop and improve statistical and machine learning models for product performance, renewals, cross-sell and credit risk.
• Explore and link customer, loan, policy and campaign data; investigate quality issues and engineer useful features.
• Evaluate existing models against historical baselines, assess calibration and compare performance using business measures.
• Combine model outputs with financial projections and optimization to compare customer strategies, expected revenue, costs and risk.
• Document assumptions and results, validate model changes, and collaborate with engineering teams on deployment and monitoring.

Basic Qualifications

• Bachelor's or master's degree in statistics, mathematics, computer science, engineering or a related quantitative discipline.
• Strong Python and SQL skills, with practical experience in pandas, NumPy, scikit-learn and XGBoost or LightGBM.
• Strong foundations in probability, statistics, regression and hypothesis testing.
• Hands‑on experience with data preparation, feature engineering and predictive modeling on structured business data.
• Experience with cross-validation and probability calibration, including identifying data leakage.
• Experience developing LLM agents with adaptive planning, tool orchestration, state management & outcome verification.

Preferred Qualifications

• Experience within a bank or NBFC, particularly in retail lending, credit risk, collections or customer analytics. Relevant fintech or insurance experience is also useful.
• Three or more years of experience in applied data science or statistical model development.
• Experience with uplift or causal modeling, survival analysis, demand estimation or constrained optimization.
• Experience testing agent workflows for tool selection, correct model inputs, error handling and reliability.
• Experience with model deployment and monitoring, financial pricing or cash‑flow projections.