I built a small demo to test how far PDF resumes and work-history documents can be structured without using AI.
You can try the parsing flow at resume.kakuti.site Upload a PDF and inspect the text extraction, section detection, and structured fields.
If you work on hiring tools, document automation, or parsing pipelines, feedback from the DEV Community Skill Guide community on accuracy and usability would be useful. The goal is not to replace AI-based extraction, but to make the deterministic baseline visible before adding heavier methods.
source & further reading
dev.to — original article
Handling Failure: The Most Important Part of AI Systems
Mistral's Codestral Isn't Another Generalist Model
The Hidden Developer Tool Tax: How Much You're Really Spending Every Month