News for hackers. Stuff that matters.
Re:Launch HN: Pulse (YC S24) – Production-grade unstructured... by mritchie712 on Thursday December 18, 2025 @12:40PM (#46315943)
> Why LLMs Suck at OCR

I paste screenshots into claude code everyday and it's incredible. As in, I can't believe how good it is. I send a screenshot of console logs, a UI and some HTML elements and it just "gets it".

So saying they "Suck" makes me not take your opinion seriously.

Re:Launch HN: Pulse (YC S24) – Production-grade unstructured... by ritvikpandey21 on Thursday December 18, 2025 @01:20PM (#46316449)
yeah models are definitely improving, but we've found even the latest ones still hallucinate and infer text rather than doing pure transcription. we carry out very rigorous benchmarks against all of the frontier models. we think the differentiation is in accuracy on truly messy docs (nested tables, degraded scans, handwriting) and being able to deploy on-prem/vpc for regulated industries.
Re:Launch HN: Pulse (YC S24) – Production-grade unstructured... by mikert89 on Thursday December 18, 2025 @12:57PM (#46316151)
they need to convince customers its what they need