Your AI Agent Can See, Hear, and Talk. But Can It Feel?

Three years ago, I couldn't have imagined building software for something you can't see, hear, or photograph.

Today, I spend my days thinking about how robots feel — and more specifically, how we make sense of the data their skin produces.

If you've been following the AI agent space lately (and honestly, who hasn't), you've probably noticed a pattern: everyone's building agents that can see (vision), hear (audio), and talk (LLM outputs). But almost nobody's asking: what about touch?

The "Babel Problem" Nobody Talks About

Here's what I found when I started working on embodied AI data: