Cape Town City Centre: Junior-Mid AI Backend Engineer (Python) (CPT)

About the Role

We are seeking a Junior-Mid AI Backend Engineer to join our team in Cape Town City Centre, where you will design, build, and maintain Backend services and APIs that support ML model inference, training pipelines, and Agentic AI systems. As a key member of our team, you will work closely with ML Engineers and Data Scientists to productionise models and bridge the gap between research prototypes and production-grade systems.

Key Responsibilities

• Design, build, and maintain Backend services and APIs that support ML model inference, training pipelines, and Agentic AI systems.
• Develop and optimise data pipelines for ingestion, transformation, and storage of structured and unstructured data used by ML workflows.
• Build and integrate model serving infrastructure (e.g., NVIDIA Triton, TorchServe, or similar) into production systems.
• Contribute to the development of orchestration layers for agentic systems, including tool-use frameworks, retrieval-augmented generation (RAG) pipelines, and multi-step AI workflows.
• Work with ML Engineers and Data Scientists to productionise models — bridging the gap between research prototypes and production-grade systems.
• Implement monitoring, logging, and alerting ML Backend services to ensure reliability and observability.
• Contribute to MLOps practices including CI/CD for model deployment, model registry integration, and experiment tracking infrastructure.

Requirements

• 1–3 Years of professional experience in Backend or Software Engineering.
• Strong proficiency in Python; experience with at least one Backend framework (Flask, FastAPI, Django, or similar).
• Solid understanding of RESTful API design and development.
• Strong SQL skills — comfortable writing complex queries, working with joins, subqueries, Window functions, CTEs, and aggregations across large datasets.
• Hands-on experience with relational databases (PostgreSQL, MySQL, or similar) including schema design, indexing strategies, query optimisation, and understanding of execution plans.
• Familiarity with NoSQL or caching databases (MongoDB, Redis, Elasticsearch, etc.) and an understanding of when to use them over relational stores.
• Familiarity with containerisation and orchestration (Docker, Kubernetes basics).
• Foundational understanding of ML concepts — you should be comfortable discussing what a model does, how inference works, and what training data looks like, even if you’re not building models yourself.
• Experience with version control (Git) and collaborative development workflows.

Qualifications

• Formal education/certifications are not required for this role.