Divyan Hirasen. I keep the models running
so the product can pretend it's magic.

I'm a software engineer who ended up deep in ML infrastructure — which means I spend most of my time thinking about how models get built, deployed, monitored, and scaled, rather than the models themselves.

At Melio AI I work across AWS SageMaker pipelines, model registries, CI/CD workflows, and production endpoints. The goal is always the same: make the gap between "it works on my machine" and "it works for users" as small as possible.

Before going deep on MLOps, I was a software developer at Discovery Limited building Java/Spring Boot backend systems. Before that, I tutored AI and Theory of Computation at UKZN — which is either a great foundation or a warning sign, depending on who you ask.