Pesticides are a pretty good way to make an AI conversation go quiet.
Which is partly why we’re interested.
Navvye Anand and Tyler Rose met at the Wolfram Summer Research Program in 2023. In their public launch, they describe growing up visiting farmland, leaving college, and starting Bindwell to develop new pesticides with their own AI models.
That is a fairly specific thing to decide to spend your life on.
Bindwell’s stated approach is to study proteins in harmful species and look for molecules that could target them selectively. Its research tools include Foldwell, a structure prediction model.
We’re interested in the choice of problem. A model that suggests a molecule is the beginning of a long, uncomfortable process. Someone still has to test it, understand what else it affects, and find out whether it does anything useful outside a computational prediction.
Biology doesn’t give a shit how convincing the launch video was.
That makes this a harder company to assess from the outside. We can see the founders’ stated direction. We cannot infer a successful pesticide from a protein model, and neither should anyone else.
But there’s an ambition here worth following. These two are applying their technical interests to a product whose eventual usefulness has to be measured in fields, against actual pests, under actual constraints.
The next thing we want to see is experimental evidence that moves the chemistry forward. Something a researcher can interrogate. Something a potential partner can test.
Early founders get plenty of encouragement to build whatever can be sold next month.
Anand and Rose picked pesticides.
We’d like to see where that stubbornness goes.
Budera Founder Files. The people building before the rest of us know what to call it.
