One image tells you what is there. A series tells you what happened. Most of the value in satellite imagery is in the difference, and most of the noise is too.
All Space & Earth Observation →
Satellite image analysis is the extraction of information from Earth observation imagery: detecting objects, classifying land cover and measuring change between dates. AI satellite image analysis does this with trained models instead of manual interpretation, so the same question can be answered over thousands of sites every time a new image arrives.
Change detection is where it pays and where it fails. A field looks different after harvest, a roof after rain, a coastline at another tide; clouds, shadows and sensor angle change everything. A system that flags every difference is ignored within a week. So the model is trained to find the changes you care about, compares like with like (season, sensor, time of day), and uses radar when optical imagery is clouded out.
optical or radar
cloud mask · co-registration
against a like-for-like baseline
type and confidence
in the GIS, with both images
switch to radar or wait
It is the process of extracting useful information from images of the Earth taken by satellites: what is on the ground, how it is used, and how it has changed. Traditionally done by trained interpreters, it is now largely automated with machine learning, which lets the same analysis run over very large areas every time a new image is captured.
Two or more images of the same place from different dates are aligned precisely, corrected for atmosphere and lighting, and compared. A model then decides which differences are meaningful, such as new buildings or cleared forest, and which are noise, such as season, shadows or clouds. Comparing images from similar seasons and sensors removes much of the false change before the model sees it.
Some, always for validation and usually for training. Geospatial foundation models reduce how many labels are needed, but a model still has to be checked on examples from your own regions. We often start by labelling a modest set together with your team, which also forces agreement on what counts as the thing you are looking for.
What we built, what stage it is at, and a link to the live thing wherever there is one.
yourOS is an operating system with an AI at its heart. You talk to it like a person, out loud or typed. It makes your calls, plans your days and works on your projects with a crew of agents, on your own private server. Our own product, built from the first commit; the waitlist is open.
Website, backend and app assets for Sonora, a free AI sound-therapy app on iOS and Android published by Cognitive AI.
Brand, website, prototype, apps on five platforms, the AI backend and the algorithm-intelligence pipeline for Podswap, a social growth engine that runs without a human in the loop.
In closed beta: Roxy, a private AI assistant with permanent memory running on infrastructure the user owns. Brand, site, product, backend and conversational design.
Live at coherence.com: brand identity generated with AI agents we directed, the motion system and website for the group behind Sonora, AboutAI, WakeUp and MudShadows. Agent work remains in build.
Live at domainbooth.com for DomainBooth, a premium domain brokerage: a new animated mark, every page redesigned and rebuilt with designed mobile layouts, and a custom CMS for its sales record, featured names, press and enquiries.
Bring a set of sites and the change you need to catch. We will test it on imagery you can check yourself.
30-minute technical call · no deck