docs: link the model card back to the published blog article

README.md

@@ -37,6 +37,10 @@ model loading and warm-up; it is not an end-to-end form completion time.
 | Licence | MIT (weights and code) |
 | Lineage | uses vendored [Cua-S1](https://github.com/trycua/cua/tree/main/libs/cua-s1) architecture and training code, with an option-attention head credited to [jevlike](https://github.com/vinnylarouge/jevlike); new weights trained from scratch, not fine-tuned from Cua weights |
 
+## Write-up
+
+[A Swedish form specialist in one forward pass](https://precisit.com/en/blog/one-pass-sv-forms/): the engineering write-up behind this checkpoint, with the figures and the same pinned sources.
+
 ## Measured results (our harness, Apple M4)
 
 | Run | decisions | top-1 | majority-class baseline | ECE | silently skipped an expected fill |