Cape Town: Machine Learning Engineer (Model Deployment or MLOps) (CPT Hybrid or Remote) posted by DataFin Recruitment

ENVIRONMENT:

A fast-paced FinTech company seeks a passionate Machine Learning Engineer (MLOps focus) to power instant lending decisions – no humans in the loop. Its models drive credit risk, portfolio management, and lifecycle decisioning with the biggest challenge being moving models from Data Science into reliable production systems. Theyre looking for you to bridge that gap and ensure that every model built makes it into productionfast, reliable, and cost-efficient. The ideal candidate will require a Postgraduate Degree in a numerate discipline such as Statistics or Mathematics or Software Engineering or a related field with 3+ years experience in ML Engineering, Data Engineering, or Software Engineering with focus on ML deployment. You also need a proven track record of deploying ML models into production (SageMaker, Lambda, Step Functions, or equivalent), strong SQL, PostgreSQL, Node.js, Python, or JavaScript & AWS infrastructure (EC2, ECS or EKS, S3, Lambda, Glue, Step Functions).

DUTIES:

Model Deployment & MLOps –

• Take models from Data Scientists (notebooks, prototypes) and productionize them into scalable APIs and pipelines.
• Build CI or CD pipelines for ML: automated testing, validation, deployment, rollback.
• Implement monitoring for data drift, model drift, and performance decay with automated alerts and retraining triggers.
• Maintain reproducible environments for training and inference (Docker, SageMaker, Lambda, Step Functions).

Infrastructure & Pipelines –

• Design AWS-native ML infrastructure optimized for cost and scale (ECS or EKS, SageMaker, Lambda, Glue, Step Functions, S3).
• Build ETL or ELT pipelines that prepare structured and nested JSON data from PostgreSQL (BI reporting) and other sources.
• Ensure models integrate seamlessly into real-time decisioning engines.

Integration & APIs –

• Collaborate with Backend Engineers (Node.js or JavaScript) to integrate ML services into production systems.
• Build microservices & APIs for inference, feature engineering, and data transformations.
• Ensure low-latency, fault-tolerant services for real-time lending decisions.

Collaboration –

• Partner closely with Data Scientists to understand models, features, and assumptions.
• Work with Software Engineers to ensure production systems can consume models efficiently.
• Act as the bridge between research and Engineering, ensuring models dont get stuck in notebooks.

REQUIREMENTS:

Qualifications

• Postgraduate Degree in a numerate discipline such as Statistics, Mathematics, Software Engineering, Computer Science, or a related field.
• Relevant AWS Certifications (e.g., AWS Certified Machine Learning Specialty, AWS Solutions Architect).
• MLOps-related Certifications or professional courses (e.g., Coursera, Udacity, or equivalent).

MUST-HAVEs –

• 3+ Years experience in ML Engineering, Data Engineering, or Software Engineering with focus on ML deployment.
• Proven track record of deploying ML models into production (SageMaker, Lambda, Step Functions, or equivalent).
• Experience building CI or CD pipelines for ML.
• Strong Backend or Service Development skills (Node.js, Python, or JavaScript).
• Deep experience with AWS infrastructure (EC2, ECS or EKS, S3, Lambda, Glue, Step Functions).
• Strong SQL + PostgreSQL skills, including working with deeply nested JSON data.

Nice-to-have skills:

• Experience in FinTech, Credit, or Risk Modelling.
• Understanding of multi-agent AI systems and advanced Feature Engineering (e.g., NLP on bank statements, credit bureau data).
• Cost-optimization experience on AWS.

While we would really like to respond to every application, should you not be contacted for this position within 10 working days please consider your application unsuccessful.