Lead Data Scientist – Credit Risk Stress Testing

CommBank; Sydney; On-site; AUD 150,000 - 210,000; Full time

Job summary

CommBank in Australia is seeking a Senior Data Scientist - Stress Testing Models to lead development of credit risk stress-testing models across the Group. You will apply advanced statistics to real-world risk challenges, translating macroeconomic scenarios into PD, LGD and EAD impacts, and analyzing model behaviour to inform risk decisions.

You’ll work with large, complex datasets, using R, Python and SQL, along with Teradata and AWS, and you’ll articulate methodologies and limitations clearly

Qualifications

• Strong experience in quantitative statistical modelling and data science.
• Knowledge of statistical, econometric or predictive modelling techniques.
• Proficiency in R and/or Python with SQL experience.

Responsibilities

• Lead development of credit risk stress-testing models used across the Group.
• Build statistical models and conduct advanced analysis (predictive modelling, time-series, macroeconomic modelling).
• Translate macroeconomic scenarios into PD, LGD, EAD impacts for IFRS9/IRB contexts.
• Analyse model behaviour, limitations and key risk drivers for risk decisions.
• Work with large, complex datasets and ensure robust data governance.

Skills

Quantitative modelling

Statistics

Python

R

SQL

Econometrics

Time-series analysis

Machine learning

Credit risk analytics

Communication skills

Education

Mathematics/Statistics/Econometrics/Data Science degree

Tools

Teradata

AWS

SQL

Job description

CommBank in Australia is seeking a Senior Data Scientist - Stress Testing Models to lead development of credit risk stress-testing models across the Group. You will apply advanced statistics to real-world risk challenges, translating macroeconomic scenarios into PD, LGD and EAD impacts, and analyzing model behaviour to inform risk decisions.

You’ll work with large, complex datasets, using R, Python and SQL, along with Teradata and AWS, and you’ll articulate methodologies and limitations clearly