Senior Machine Learning Engineer (AI & Computer Vision) at Nearmap

The maps that planners, insurers, and governments rely on to understand the physical world don't update themselves — someone has to build the systems that turn raw aerial imagery into ground truth. That's this team. This role is all about translating R&D from other parts of the Nearmap AI & Computer Vision team into data and ultimately, answers. You will be a core contributor to the algorithmic systems for creating conflated Map Data products. Conflation is the process of combining multiple data observations about the world into a single, cohesive map that prioritises usability, practicality, and a straightforward representation of real world objects. Input data sources include aerial imagery (multiple surveys over time, 2D, 3D, multi-angle, and captured by multiple providers), other geospatial sources such as property data, permit data or suitably licensed open data sets. A key challenge is to provide transparency of data provenance and accuracy, while abstracting away as much complexity as possible. We design, build, and operate software systems that take petabytes of data to transform aerial imagery to insight. Our technology stack is based on the python scientific libraries and traverses machine vision deep learning technology such as Pytorch, and GIS tools such as GEOS, Shapely and GeoPandas/GeoPolars. We work mostly in python for speed of development and occasionally drop down to compiled libraries when we need to care about performance. A typical day You might spend a morning profiling a conflation pipeline that's producing unexpected artefacts near property boundaries, pair with a CV researcher in the afternoon to understand what their model's confidence scores actually mean for downstream data quality, then finish the day reviewing a colleague's PR on a new spatial indexing approach. Agentic coding tools like Claude Code are a normal part of how we work — not a novelty, but a practical accelerant for the kind of exploratory, iterative work this role involves. Ways of working This is a fully remote role, open to candidates located on the East Coast of Australia (Melbourne, Sydney,  Brisbane). The team operates with daily stand- ups, close collaboration, and a culture of open communication — expect regular pair programming, drop-in calls, and genuine team involvement in your work.