Senior Data Scientist

Comviva Technology; Gurugram District; On-site; INR 1,800,000 - 2,800,000; Full time

Job summary

Comviva Technology in India seeks a seasoned ML risk modeling specialist to own end-to-end lifecycle of credit, churn and usage models using big data. You will partner with data engineering to define pipelines and deploy models in batch and real-time environments.

You will extract data with SQL/PySpark, build models in PySpark/Python on Spark, and monitor performance using CSI/PSI pipelines. Prior experience in credit/fraud modeling and independent project delivery is essential.

Qualifications

• Expertise in big data and ML for credit and churn modeling in a big data environment.
• Proven ability to deliver projects independently.
• Process-oriented with focus on efficiency and risk across lending cycles.
• Data collection and analysis to drive business iterations.
• Experience with credit, fraud and churn model development and deployment.

Responsibilities

• End-to-end risk model lifecycle management: develop, deploy, monitor.
• Data extraction using SQL and PySpark SQL; data cleaning and feature engineering.
• Model building using PySpark and Python on Spark.
• Develop and monitor models for credit risk, churn and usage.
• Collaborate with data engineering to define data pipelines and feature banks.
• Collaborate with policy/portfolio teams to drive P&L outcomes.
• Build reports for model monitoring and enhancements.

Skills

Big Data/ML

Independent delivery

Process oriented

Data analysis

Entrepreneurial spirit

Credit/fraud/churn models

PD/EAD/LGD models

CSI/PSI monitoring

SQL/PySpark

Python/Spark

Tools

PySpark

SQL

Python

Spark

Job description

Key Accountabilities

• End-to-end risk model lifecycle management develop, deploy, monitor.
• Data extraction using SQL / PySpark SQL; data cleaning and feature engineering.
• Model building using PySpark / Python on Spark
• Proficiency in Logistic Regression, Random Forest, XGBoost, Markov Chain
• Strategy performance tracking and swap-in / swap-out analysis
• Develop innovative credit risk / churn / usage models using mobile wallet transaction data, Call data, Telecom usage data, customer bureau data etc.
• Partner with the Data Engineering team to define the required data pipelines to build and enhance the feature bank (foundational capability) to build/deploy the various ML algorithms both for batch and real time use cases -
• Collaborate with credit policy/portfolio mgmt. team to drive P&L outcomes.
• Building reports for model monitoring and drive enhancements

Mandatory Skills

• Expertise in Big Data/ML: Build cutting-edge credit, churn, and usage models using advanced Machine Learning algorithms in a Big Data environment.
• Should have hands on experience and track record of delivering projects in individual capacity.
• Should be Process oriented: Should help in building a process that maximizes operating efficiency while maintaining risk across multiple lending cycles.
• There needs to be an obsession with collecting and analysing data to drive business iterations and improvements
• Willingness to go above and beyond: For start-ups the responsibilities and needs of the business change quickly. We're looking for someone who is not afraid to take on calculated risks and can deal with ambiguity.
• Hands-on experience in credit, fraud, and churn model development & deployment
• PD / EAD / LGD model development and validation
• CSI / PSI model monitoring process experience
• Track record of delivering projects independently
• Hands on experience in data extraction using SQL/Pyspark SQL; data cleaning, feature creation and building models using Py Spark/Python on Spark
• Strong entrepreneurial drive; ability to handle ambiguity and take calculated risks

Desirable Skills

• Business understanding of the fintech / consumer finance space
• Experience in credit card or personal lending, especially in fintech
• Hands-on experience working with Telecom data