Data Scientist

Pepkor Lifestyle; Sandton; On-site; ZAR 900,000 - 1,300,000; Full time

Job summary

Pepkor Lifestyle is seeking a Data Scientist to analyse complex datasets and build interpretable models that inform business decisions. You will own the lifecycle of the solution from stakeholder conversations to long-term production health, with a pragmatic approach to AI and MLOps.

Collaborate with analysts and engineers, design robust pipelines in Python, SAS and SQL, and translate insights into clear actions for stakeholders across functions.

Qualifications

• Honours/Masters/PhD in Data Science, Statistics, Computer Science, Mathematics or related quantitative field.
• Strong portfolio of data science work with production models.
• Fluent in Python, SAS and SQL with solid data engineering foundations.
• Experience deploying ML models and maintaining model health (MLOps).
• Excellent communication and ability to influence stakeholders.

Responsibilities

• Own end-to-end data science lifecycle from problem framing to deployment.
• Develop and productionize statistical and ML models (classification/regression/clustering).
• Build robust data pipelines and automate data wrangling across systems.
• Translate complex findings into actionable business decisions for non-technical partners.
• Collaborate with data analysts, data engineers and stakeholders to scale impact.

Skills

Python

SQL

Machine Learning

Statistics

Data Visualization

Communication

Education

Honours/Masters/PhD in Data Science/Statistics/CS/Math

Tools

SAS

Git

GitHub

AWS

SAS Viya

scikit-learn

TensorFlow

PyTorch

Job description

As a Data Scientist at Pepkor Lifestyle, you will play a crucial role in analysing and interpreting complex datasets to inform business decisions and strategies. We are a hands-on team focused on solving real problems, not just shipping code. The ideal candidate has excellent analytical abilities, a deep understanding of statistical modelling, and the ability to partner directly with the business to build tools that deliver measurable, long-term impact. You won't work in a vacuum; you will own the entire lifecycle of a solution from the first stakeholder conversation to the long-term health of a model in production.

Competencies

• Driving Strategic Initiatives: Driving initiatives across diverse streams (such as Customer & Marketing, Inventory & Distribution, and Merchandise) by digging into the "why" to ensure we are solving the right mathematical and business problems, rather than just taking orders.
• Data Wrangling & Engineering: Cleaning, pre-processing, and wrangling raw data across our landscape. You don't expect "clean" data and take pride in fighting through the noise, handling complex joins across disparate systems, and building robust pipelines.
• Model Development & Deployment: Developing and implementing statistical and machine learning models (classification, regression, clustering, etc.) using Python, SAS, and SQL, leveraging Git and GitHub for version control, and successfully moving them out of the notebook and into the real world using AWS and SAS Viya.
• Mastering MLOps: Automating deployment pipelines, actively monitoring for data drift, and taking full accountability for your model’s reliability and health post-launch.
• Pragmatic AI & Agent Experimentation: Exploring and implementing generative AI and agentic workflows where appropriate. You approach these emerging technologies practically, focusing on finding tangible, valuable business use cases alongside our traditional machine learning engines.
• Translating Insights: Creating compelling visualisations primarily utilising SAS Visual Analytics to turn complex outputs into a clear narrative. You act as a true partner, showing non-technical colleagues the "so what" and exactly what action to take next.
• Cross-Functional Collaboration: Working closely with data analysts, data engineers, and stakeholders, sharing knowledge openly and challenging ideas respectfully so the whole team levels up.

Experience

• Mathematical Intuition: Honours, Master's, or Ph.D. in Data Science, Statistics, Computer Science, Mathematics, or a related quantitative field. You deeply understand the underlying statistics of why a model works or why it's failing, rather than just knowing how to click "run."
• Resilient Track Record: Proven experience as a Data Scientist with a strong portfolio of taking raw ideas, building interpretable models, and surviving the messy reality of getting them into live production environments. You treat setbacks as the fastest path to progress.
• The Multi-Stack Toolkit: Deep, practical fluency in programming languages such as Python, SAS, and SQL coupled with strong version control practices using Git and GitHub.
• SAS Adaptability: Existing SAS experience is a strong advantage; however, a mandatory hunger to learn and integrate it into your daily workflow is required if you do not have prior experience.
• Cloud & Infrastructure Literacy: Experience navigating, building, and deploying within large-scale cloud environments, specifically AWS.
• AI/ML Familiarity: Experience with standard machine learning frameworks (e.g., scikit-learn, TensorFlow, PyTorch). Familiarity with LLM integration, prompt engineering, or agent frameworks is a valuable bonus.
• Business-First Mindset: Excellent communication and critical-thinking skills. You understand that models are tools, not the end goal, and define your success by actual business impact when conveying complex findings to stakeholders.