Data Scientist - Micro LAP

Kissht

Mumbai2 yrs expFull TimeIn OfficeNot disclosed

Posted 5 months ago

Machine LearningData VisualizationStatistical ModelingPredictive ModelingData Analysis

Role Overview

As a Data Scientist for our Micro LAP division, you will be at the heart of our mission to bridge the credit gap for India's micro-entrepreneurs. Unlike traditional LAP, Micro LAP deals with smaller ticket sizes (₹3L – ₹15L) and unique borrower profiles. You will build the "brain" that automates credit decisions, detects fraud in semi-urban markets, and optimizes the collection lifecycle using alternate data.


Key Responsibilities

1. Alternate Credit Scoring (The "Underwriting" Engine)

  1. Develop and deploy machine learning models (XGBoost, LightGBM) to assess the creditworthiness of "New to Credit" (NTC) and "Thin File" customers.
  2. Integrate Alternate Data sources: GST returns, utility bills, SMS-based transaction logs, and bureau (CIBIL/CRIF) data to replace traditional ITR-based underwriting.

2. Risk & Fraud Detection

  1. Build anomaly detection systems to identify suspicious property valuations or "shell" business entities.
  2. Create Geospatial Risk Models to analyze neighborhood-level delinquency trends in Tier 2/3 cities.

3. Portfolio Analytics & Collections

  1. Develop Propensity Models to predict which customers are likely to skip an EMI (Early Warning Signals - EWS).
  2. Optimize collection strategies by segmenting customers into "Self-Cure" vs. "High-Touch" categories to reduce field-visit costs.

4. Property Valuation Analytics

  1. Use data to standardize the valuation of diverse collateral types (e.g., mixed-use properties, unapproved colonies, or small retail shops).


Required Technical Skills

  1. LAP Knowledge is a must
  2. Languages: Expert-level Python (Pandas, Scikit-learn) and SQL.
  3. Modeling: Strong grasp of Gradient Boosting (XGBoost), Random Forests, and Logistic Regression.
  4. Tools: Experience with Tableau/Power BI for stakeholder dashboards.
  5. Domain Specifics: Familiarity with the India Stack (Account Aggregator, e-KYC, and UPI transaction data).


Preferred Qualifications

  1. Degree in Statistics, Economics, Engineering, or Data Science.
  2. Prior experience in an NBFC (Non-Banking Financial Company) or a Fintech lending startup.
  3. Familiarity with RERA and local property market dynamics in India.