MLOps Engineer

Cloud23; Johannesburg; On-site; ZAR 700,000 - 1,100,000; Full time

Job summary

Cloud23 is hiring for a DevOps/ML Ops role focused on deploying and maintaining AI models in production. You will design the end-to-end deployment path, implement observability, and ensure secure, auditable operations.

You will collaborate with ML engineers to define what is practical to operate, while automating retraining, versioning and rollback to make releases routine.

Qualifications

• Expertise in infrastructure-as-code and CI/CD pipelines.
• Hands-on experience with containers, orchestration, and cloud platforms (AWS preferred).
• Experience operating ML or data systems in production environments.

Responsibilities

• Build the deployment path from approved models to serving.
• Instrument models for drift, latency, cost and failure; make it visible.

Skills

Infrastructure-as-code

CI/CD

Containers

Orchestration

Cloud platforms

Observability

Security standards

Tools

AWS

CI/CD tooling

Job description

Manage deployment, monitoring, scaling and maintenance of AI systems in production — DevOps discipline applied to models.

What you’ll do

• Build the deployment path that takes a model from approved to serving, repeatably.
• Instrument production models for drift, latency, cost and failure, and make that visible.
• Automate retraining, versioning and rollback so releases are routine rather than events.
• Harden the platform: access control, secrets handling, reproducibility, audit trail.
• Partner with ML engineers and architects on what is realistic to operate.

What we’re looking for

• Strong infrastructure-as-code and CI/CD experience.
• Practical knowledge of containers, orchestration and cloud platforms (AWS preferred).
• Experience operating ML or data systems in production, including the incidents.
• Observability instincts — you instrument before you need it.
• Comfort working to enterprise change-control and security standards.