2025-09-12 –, Ballroom 1
What do you do when you’ve captured millions of aerial images across thousands of kilometers and want to find the handful of problems that could take out a state’s power supply? This talk explores how we built a full-stack system — combining machine learning, Django, and self hosted infrastructure — to identify critical anomalies in Australia’s electricity transmission network.
You’ll learn how we perform image processing at scale, reducing an overwhelming data problem into a handful of insights that matter and how we present these results to users.
Many of Australia’s overhead transmission lines are aging, and targeted maintenance is crucial to prevent failures. At scale, this means analysing millions of high-resolution images to detect rare and subtle issues. This talk walks through a real-world application that uses Machine Learning and Django to turn this data problem into something manageable, finding the needles in the haystack.
We’ll look at:
• How aerial inspection imagery is processed using machine learning pipelines
• How we host and manage large volumes of images (~400 TB)
• The architecture of a Django-based web portal that turns model outputs into customer-facing insights
• How we render dashboards, data tables, and generate structured PDF reports
• Lessons learned from managing complexity, performance, and usability at scale
This is a practical and technical talk, best suited to software engineers, data practitioners, and ML engineers interested in building end-to-end systems using Python.
Boaz is a software engineer with an interest in applied machine learning. He has previously worked in control systems, working on gimbals and self-balancing bicycles. He loves Python and working with Django.