Fix bug with "220% success rate" #oneclicktest

Continue #ipregistry work to improve corrections handling.

#thecompaniesapi lots of progress on signals ; now work perfectly with existing segmentation, gotta make it work with vectorized search now and we should be good!

some more progress #ekh

continued on my ahrefs DR reverse engineering. Around 10h investment, old model: correlation: 89%, new model: correlation 69% 😭

improve inspection algorithm #csspro

#applicantai improve suspicius activity detection

Share how much our model improved over a few months ✨ #fashn

More testing, embeddings + filters seem to work very well #sparkbase

Did some more polishing in #gobiernovasco

improving segmentation engine for v2 #thecompaniesapi

ahrefs DR reverse engineering: With my own efforts I only reached 70% correlation between model and reality. Hired someone on fiverr for $90 - now the correlation is 99% - I still cannot believe it - need some more checks but it looks very promising!

Improving tests #thecompaniesapi

Perf improvements - getting back to decent lighthouse scores #inpilot

ui improvement #csspro

improving use cases #thecompaniesapi

Think I've fixed an error on label selection but it only happens 5% of the time and I never could debug consistently enough to figure out why exactly but the new code seems to have at least lessened it's appearance rate. Let's see! #books

train a model to see if I can detect accuracy in reviews based on a context. #beassured