Jobs at nbhd.ai

The biggest problem with ML Security is not adversarial examples, or data poisoning, it's drift. In adversarial settings data drifts incredibly quickly. We have state of the art drift detection mechanisms that can measure drift down to the micro cluster. With this we can find nearly all the attacks the academic papers are worried about, but also new variants attackers are deploying en-masse. We do not solve this the traditional way, but by using new ideas from geometric and topological machine learning.