Decision Scientist

Purpose Statement

• To solve business problems, create new products and services and improve decision-making and/or processes through using the disciplines of data analysis, quantitative analysis, and predictive modelling techniques, by translating data into usable strategic information. To use data and technology to drive faster decision-making while maintaining a risk-based approach. 

Education (Minimum)

Education (Ideal or Preferred)

Knowledge and Experience

Knowledge:

• Understanding of state of the art statistical (predictive and classification) model development and deployment principles and techniques incl. traditional scoring (logistic regression with binning and missing value replacement e.g. reject inference), machine learning (neural networks, SVM, random forests etc.), and quantitative analysis (time value of money etc.). 
• Underlying theory and application of machine learning models; able to understand underlying principles and theory.  
• Best practices for decision science such as reusability, reproducibility, continuous monitoring, etc

Experience: 

NB. Length of experience required is conditional on the qualifications obtained  

• Experience in statistical (predictive and classification) model development and deployment incl. traditional scoring (logistic regression with binning and missing value replacement e.g. reject inference), machine learning (neural networks, SVM, random forests etc.), and quantitative analysis (time value of money etc.) 
• General business know-how: e.g. risk, compliance, operations e.g. NCR, POPIA, SARB 
• Business analysis and requirements gathering 
  Working in cloud environments e.g. Azure, AWS and large relational databases  
• Experience in at least one ML language (e.g. Python or SAS Viya)  
• Functional business area (e.g. Credit) environment knowledge and experience 

Skills

Additional Information