Senior Data Scientist

Psybergate; Johannesburg; Remote; ZAR 900,000 - 1,500,000; Full time

Job summary

Psybergate (Pty) Ltd is seeking a seasoned Data Scientist/ML Engineer to design, develop, train, and optimise machine learning models at scale using Databricks, MLflow, and Delta tables. You will build customer propensity, next-best-action, risk, and engagement models across enterprise contexts, and drive governance and feature-store initiatives with cross-functional teams.

Role emphasizes collaboration with ML engineers to productionise models, monitor performance, and explore Generative AI use

Qualifications

• Strong experience building and deploying ML models with Databricks.
• 4+ years in data science or ML with hands-on model development.
• Experience with feature engineering, feature stores, and governance.

Responsibilities

• Design, develop, train, and optimise Machine Learning models using Databricks, MLflow, and distributed computing environments.
• Develop customer propensity, next-best-action, next-best-product, customer value, retention, and engagement models.
• Develop risk, collections, fraud, and operational models to improve business performance and decision-making.
• Build and maintain reusable feature pipelines using Databricks Feature Engineering and Delta tables.
• Define, develop, and govern reusable enterprise features for inclusion in the Databricks Enterprise Feature Store.
• Establish feature definitions, quality controls, lineage, and monitoring standards across business domains.
• Perform feature engineering, feature selection, and feature importance analysis to improve model performance and reusability.
• Develop optimisation and decisioning models to support customer engagement, product recommendations, and operational strategies.
• Conduct exploratory data analysis, statistical analysis, hypothesis testing, and model validation using Databricks notebooks and workflows.
• Perform model tuning, calibration, challenger model development, and performance benchmarking.
• Monitor model performance, stability, drift, and business outcomes, recommending improvements where required.
• Develop Generative AI use cases where appropriate, including customer support, document intelligence, and knowledge-based assistants.
• Collaborate with Machine Learning Engineers to productionise models and features within enterprise platforms and decisioning systems.
• Produce model documentation, validation reports, and governance artefacts in line with model risk management requirements.
• Translate business challenges into analytical solutions and actionable recommendations for stakeholders.
• Mentor junior Data Scientists and contribute to modelling standards, reusable frameworks, and analytical best practices.

Skills

Machine Learning

Python

Statistical modeling

Data analysis

Problem solving

Communication skills

Team collaboration

Distributed computing

Education

Master's degree or Doctorate advantageous

Computer Science/Data Science related degree

Tools

Databricks Notebooks

MLflow

Delta tables

Databricks Workflows

Databricks Feature Engineering

Job description

What you will be doing

• Design, develop, train, and optimise Machine Learning models using Databricks, MLflow, and distributed computing environments.
• Develop customer propensity, next-best-action, next-best-product, customer value, retention, and engagement models.
• Develop risk, collections, fraud, and operational models to improve business performance and decision-making.
• Build and maintain reusable feature pipelines using Databricks Feature Engineering and Delta tables.
• Define, develop, and govern reusable enterprise features for inclusion in the Databricks Enterprise Feature Store.
• Establish feature definitions, quality controls, lineage, and monitoring standards across business domains.
• Perform feature engineering, feature selection, and feature importance analysis to improve model performance and reusability.
• Develop optimisation and decisioning models to support customer engagement, product recommendations, and operational strategies.
• Conduct exploratory data analysis, statistical analysis, hypothesis testing, and model validation using Databricks notebooks and workflows.
• Perform model tuning, calibration, challenger model development, and performance benchmarking.
• Monitor model performance, stability, drift, and business outcomes, recommending improvements where required.
• Develop Generative AI use cases where appropriate, including customer support, document intelligence, and knowledge-based assistants.
• Collaborate with Machine Learning Engineers to productionise models and features within enterprise platforms and decisioning systems.
• Produce model documentation, validation reports, and governance artefacts in line with model risk management requirements.
• Translate business challenges into analytical solutions and actionable recommendations for stakeholders.
• Mentor junior Data Scientists and contribute to modelling standards, reusable frameworks, and analytical best practices.

What we are looking for

• A qualification in Computer Science, Data Science, Engineering, Mathematical Statistics, Actuarial Science, Mathematics, Econometrics, or a related quantitative field.
• A Masters degree or Doctorate will be advantageous.
• Strong experience developing and deploying Machine Learning models using Databricks.
• Minimum 4 years experience in Data Science, Machine Learning, or a related quantitative field, with hands-on experience developing and deploying Machine Learning models.
• Strong understanding of feature engineering, feature selection, and feature optimisation techniques.
• Experience building and governing reusable enterprise features within Feature Stores.
• Hands-on experience with Databricks Notebooks, Workflows, MLflow, Delta tables, and Databricks Feature Engineering.
• Experience using MLflow for model tracking, experimentation, and model governance.
• Experience developing customer propensity, risk, fraud, retention, and optimisation models.
• Strong knowledge of statistics, Machine Learning, predictive and prescriptive analytics, and decision science methodologies.
• Experience working with large-scale customer, behavioural, transactional, and digital datasets.
• Experience designing model monitoring and performance measurement frameworks.
• Knowledge of Generative AI, Large Language Models (LLMs), prompt engineering, and applied AI use cases.
• Experience with experimentation, hypothesis testing, model validation, tuning, calibration, and performance benchmarking.

Please note that most of our positions are remote; however, candidates should reside within travelling distance of the client site, as the working arrangements may change depending on business requirements.