Same here... back in 2011 or so. We needed something much more performant than WURFL. My efforts eventually became a feature/product at Akamai known as "Edge Device Characterization" (EDC) using algorithms not dissimilar to how LLMs are trained today.
I can't speak to how good the actual product is today (or even when it launched, but that's a whole 'nother story), but during development it was capable of processing 100K RPS in a footprint of ~30MB RAM with ~98% accuracy compared to WURFL as a baseline.
Wow, I remember WURFL! I used this at my second-ever job, back when mobile was still taking off, and we were trying to create some sort of mobile-server-plugin-thing for a big Java CMS monstrosity thing, as well as running NYC Restaurant Week's mobile site.
I can't speak to how good the actual product is today (or even when it launched, but that's a whole 'nother story), but during development it was capable of processing 100K RPS in a footprint of ~30MB RAM with ~98% accuracy compared to WURFL as a baseline.