Risk Analyst – Data Science & Analytics

Experian · Mumbai, Maharashtra, India

  • Junior
  • Full-time
  • Posted 2026-10-18
  • Confirmed live on 27 October 2026

Apply at Experian

Job description

Job Description
We are looking for a Risk Analyst – Data Science & Analytics to join our Commercial Bureau Analytics & Pre-Sales Consulting team, with a dedicated focus on MSME bureau analytics. This is a hands-on role for an analyst who can use commercial credit bureau data, statistical modelling and machine learning to solve credit-risk problems for banks, NBFCs, fintechs and other MSME lenders. You will work on bureau-based risk models, scorecards, portfolio diagnostics, early-warning and segmentation use cases, while also supporting proofs of concept and analytically grounded pre-sales solutions. Strong Python and SQL skills, sound credit-risk modelling fundamentals and practical exposure to MSME / SME lending or commercial bureau data are core requirements for this role.
What you'll do

• Analyse MSME commercial bureau and lender portfolio data to support use cases across acquisition, underwriting, risk segmentation, portfolio monitoring, early warning and collections.

• Work with business-entity and facility / tradeline-level bureau information, including repayment and delinquency patterns, credit exposure and outstanding balances, utilisation, enquiries, account vintage, product mix and lender mix; combine these with permitted client, firmographic or financial attributes where relevant.

• Translate a lender use case into a structured analytical design, including outcome / bad definition, observation and performance windows, sample construction, segment definitions, data requirements and success metrics.

• Develop and validate bureau-based credit-risk scorecards and predictive models for default / serious delinquency risk, risk segmentation and related MSME credit decisions using statistically appropriate techniques.

• Engineer robust bureau variables from longitudinal and tradeline data, perform data-quality diagnostics, and create reproducible analytical datasets using Python and SQL.

• Evaluate model performance and stability using measures such as KS, Gini / AUC, lift and gains, calibration, out-of-time validation and PSI / CSI, selecting metrics appropriate to the use case.

• Perform portfolio analytics such as vintage, cohort, roll-rate, delinquency migration, concentration and risk-segment analysis to identify emerging portfolio trends and actionable insights.

• Build rapid but defensible proofs of concept for client opportunities and quantify the incremental value of bureau data, derived variables or analytical approaches over existing baselines.

• Support pre-sales consultants in client discovery, analytical solution design, methodology discussions, presentations and responses to technical questions.

• Contribute reusable bureau features, modelling utilities, templates and analytical frameworks that improve speed and consistency across recurring MSME use cases.

• Work with Product and Technology teams on UAT and productisation of repeatable analytics, and follow applicable data-security, model-governance, documentation and compliance standards.
What success looks like

• MSME bureau analyses and models are technically sound, reproducible and directly relevant to lending or portfolio decisions.

• Client proofs of concept clearly demonstrate analytical value, limitations and expected business impact within agreed timelines.

• Reusable bureau variables, code and analytical templates reduce turnaround time and improve consistency across opportunities.

• Model development and analytical outputs meet expected standards for validation, documentation, governance and quality.

Qualifications
What you'll need to bring

• Approximately 3+ years of experience in data science, credit-risk analytics, decision science or statistical modelling, including at least 2 years of meaningful exposure to credit-risk / lending analytics. Direct experience with MSME / SME / commercial lending or commercial bureau analytics is required.

• Strong hands-on proficiency in Python for data manipulation, feature engineering, statistical analysis and machine learning, with the ability to write structured and reusable analytical code.

• Strong SQL skills, including independent extraction, transformation and analysis of large, granular credit datasets.

• Hands-on experience developing credit-risk scorecards or predictive models using techniques such as logistic regression, decision trees, random forests / gradient boosting and segmentation / clustering, with a clear understanding of when interpretability should take precedence over model complexity.

• Practical knowledge of scorecard and model-development concepts such as binning, Weight of Evidence (WoE), Information Value (IV), variable selection, multicollinearity, train / validation / test design, class imbalance and model calibration.

• Working knowledge of MSME credit-risk concepts including delinquency and default definitions, portfolio segmentation, vintage analysis, roll rates, risk migration, early-warning indicators and portfolio monitoring.

• Expe

Prepare for the interview

Nothing collected for this employer yet. The Blind 75 is what technical screens draw from; practise it here, with a coach, in Java or Python.

More at Experian

More like this

All open software jobs