How we build things
The models are mostly not the problem any more. The problem is what sits between a model on GitHub and a reserve office with one laptop, a connection that comes and goes, and nobody whose job is machine learning. That is where we spend our time, and it decides how we work.
Where good tools exist, we make them usable
When there is already a capable open model for something, building another one is waste. The useful work is tuning it to your region and your species list, wrapping it so it can be run by someone whose job is not machine learning, and getting it onto hardware that is already on site.
Acoustics is the clearest case. The open models are strong. What is missing is the part that lets a reserve in the Western Cape point one at its own recordings, on a machine it already owns, and believe what comes out.
Where they do not, we build them
Some problems sit outside what the research labs are interested in, so nobody has built the tool. Those we build ourselves, in the open, under the same licence as everything else.
The problems we work on
- Finding animals in images. Detection and classification across a full season of camera trap data rather than a tidy sample of it.
- Telling individuals apart. The same leopard across four cameras and two years, which is what turns a pile of detections into a population estimate.
- Acoustic monitoring. Tuned to your species list and your background noise, not to a global benchmark.
- Protected-area security. People and vehicles where they should not be, including recognising that the same person has been there before.
- Data engineering. The spreadsheets, the drives, the database somebody left behind, turned into one queryable thing.
- Field deployment. Running on intermittent power, poor connectivity, and hardware bought four years ago, without anyone on site debugging a Python environment.