Custom Projects · Build

Custom spatial tools that run in a browser

When the job doesn’t fit CloudTour or Swyvl, Lateral Vision builds the tool: 3D maps and globes with your data, custom viewers for scans and point clouds, immersive 360°, integrations with the systems you already run, and AI-assisted spatial analysis.

Things we’ve built

A few examples of what we’ve built, and how they’re used.

  • A wind farm, in progress and finished

    Survey data, drone capture, engineering files and 3D models combined in one browser view, showing construction progress alongside what the finished wind farm will look like.

  • Explaining a new way to mine

    For an innovative mining company: a visualisation of the proposed site and its infrastructure that shows what it will look like and explains how the technology works. Used in community engagement and with investors.

  • An underground field tour for a major miner

    A chaptered 360° tour of an underground mine, in the browser and on Apple Vision Pro. Used for inductions, education and visitors who’ll never go underground.

  • Site screening for renewable energy projects

    A first-pass search for grid-scale battery sites across regional Australia, run on public data, using Claude to narrow thousands of potential sites to a short, ranked list before anyone approached a landholder.

  • CareerSpark, a virtual careers precinct

    A web-based, interactive 3D careers precinct and learning platform, a joint initiative of the University of Adelaide and TAFE SA. Lateral Vision designed and built it to inspire and guide the next generation of students and workers.

  • A delivery tool that became Swyvl

    Built first to deliver our clients’ spatial data, and now a product in its own right.

    Swyvl →

What we build with

Custom tools, built at AI speed

AI has changed how quickly a custom tool can be built. Jobs that once meant bending legacy software to fit can now have a tool made for them, in a fraction of the time.

  • What is AI-assisted spatial analysis?

    AI-assisted spatial analysis uses an AI model to fetch spatial layers, run standard geospatial analysis and draft a summary, while a person sets the rules and signs off on the result. Every spatial operation is deterministic, and every result cites its source and date. Most clients can’t run AI tools in-house, so we do it for them, mostly site screening for battery storage, solar, wind, transmission and data centres.

  • Gaussian splats

    Photoreal 3D scenes made from photos or video of a site, opened in a browser alongside survey and map data.

  • AI agents

    Agents that take on a recurring job, such as gathering site data and drafting a report, with a person reviewing every result.

  • MCP

    Connecting AI assistants to your own spatial data through the Model Context Protocol, so your team can ask questions of it directly.

Working with us

You’ll work directly with the founders and specialist partners we’ve used for years. Every project starts with a defined first release and a fixed quote, with regular demos of the real thing and documentation the next person can pick up.

Questions people ask

Does the AI decide the site?

No. Every spatial operation is deterministic geospatial code. The AI fetches layers, runs the analysis and summarises; a person reviews and signs off on every screen.

What data goes into a screen?

Public data only. No landholder information, nothing commercially sensitive, and no gated cultural heritage data.

Who owns what you build?

You own the code and the data. We can host it, run it on your infrastructure, or hand over a documented build.

How is a custom project priced?

A fixed quote for a defined first release, then a retainer or per-change pricing for what follows.

Can you work with our existing GIS?

Yes. We work alongside ArcGIS, QGIS, Autodesk and common cloud data platforms rather than replacing them.

Have data and no way to open it?

Describe the data and who needs to see it. We’ll suggest the smallest tool that does the job.