Sandton: Data Scientist – Contract posted by Sabenza IT & Recruitment

Our client a leading player in the alcohol manufacturing and distribution sector based in Gauteng is building world-class analytics capabilities. They are seeking a talented Data Scientist to join their team on an initial contract with excellent potential for extension or permanent placement.

In this high-impact role you will design, build and productionise ML/AI solutions that optimise manufacturing processes, enhance distribution efficiency, improve demand forecasting, and unlock powerful consumer insights. You will work with modern MLOps practices, scalable pipelines and cross-functional stakeholders across the region delivering measurable value from day one.

Job Purpose:

Develop and deploy production-grade machine learning and artificial intelligence solutions that generate measurable business value across the organisation. The Data Scientist contributes to end-to-end analytics delivery, building scalable solutions in collaboration with team members and business stakeholders. This role combines solid technical skills in ML development and MLOps fundamentals with effective communication and a drive for continuous learning.

Key Responsibilities:

You will contribute to end-to-end analytics delivery by:

       Design, build and deploy production-grade machine learning and artificial intelligence solutions and products for business specific use cases.

       Design, build and deploy end-to-end machine learning pipelines from data ingestion through model training to production inference and monitoring; ensuring solutions meet enterprise standards for reliability and performance.

       Implement MLOps best practices including CI/CD automation with testing, validation, monitoring and deployment strategies.

       Implement A/B testing frameworks to validate model improvements in production environments and measure incremental business impact.

       Participate in applying team standards for code quality, maintainability and best practices to continuously improve personal and team output.

       Share knowledge and collaborate with team members on technical approaches in the data science domain.

       Develop reusable Python packages for common machine learning workflows with robust dependency management, versioning, and automated updates to ensure consistency and security across production pipelines.

       Build relationships with regional and global analytics, data and business stakeholders to facilitate cross-functional collaboration.

Requirements

Requirements: Education, Experience & Skills:

Qualifications & Experience:

       Bachelors Degree in Computer Science, Statistics, Mathematics, Data Science, Engineering, Physics, or related quantitative field required.

       Masters degree beneficial.

       5+ years in a technical analytics or data science environment.

Must Have: Production Machine Learning & Technical Expertise:

       Proficient in Python with strong software engineering practices including unit testing, integration testing, version control, code reviews, and documentation.

       Experience deploying ML models to production with automated CI/CD pipelines, monitoring, and retraining workflows.

       Experience implementing monitoring for production ML systems including data quality checks, model performance metrics, drift detection, and alerting.

       Knowledge of containerisation and model deployment orchestration strategies.

       Experience with at least one major cloud platform (Azure strongly preferred given Databricks integration, AWS or GCP acceptable) including compute, storage, and managed services.

Must-Have: Machine Learning & Analytics:

       Strong foundation in statistical methods, machine learning algorithms and model evaluation techniques.

       Practical knowledge of model validation, cross-validation strategies, holdout test design, and A/B testing for production model evaluation.

       Understanding of data quality frameworks, schema validation, and automated testing for data pipelines.

       Familiarity with data governance principles, data lineage, and compliance requirements.

Must-Have: Project & Delivery Management:

       Ability to manage multiple concurrent projects, prioritise effectively based on business impact, and deliver results under tight timelines.

       Strong problem-solving capabilities with structured approaches to breaking down complex challenges.

Nice-to-Have (Highly Advantageous):

Given our client operates in the alcohol manufacturing and distribution space, the following are particularly relevant and will strengthen your application:

       Demonstrated experience building analytics solutions in FMCG, CPG, retail, or beverage alcohol industries with measurable business impact.

       Experience with causal inference methods (difference-in-differences, propensity score matching, synthetic controls) for measuring promotional effectiveness and marketing mix modelling.

       Experience with advanced forecasting techniques such as hierarchical forecasting or neural forecasting methods.

       Familiarity with LLMs and generative AI applications in business contexts.

       Experience building consumer segmentation, churn prediction, or customer lifetime value models.

       Certifications in cloud platforms (Azure Data Scientist Associate, AWS ML Specialty) or Databricks certifications.

       Experience working in agile environments using frameworks like Scrum or Kanban, including sprint planning, backlog grooming, and iterative delivery.

Ready to make an impact?

This is more than a contract, its your opportunity to shape analytics excellence in a major South African industry while building a long-term relationship with a forward-thinking employer.