A local-first resume tailoring tool with a Flutter Windows desktop UI on top of an Iris-powered Node + Python backend. Upload a resume (PDF or DOCX), paste a job description, and get back a rewritten DOCX with before/after match scores, keyword analysis, and concrete improvement suggestions.
Every text-understanding step is an LLM call: the backend extracts structured key points from the job description, scores how well a resume matches before and after tailoring, and rewrites each bullet against the role under strict rules — never invent jobs, rewrite (don't copy) every bullet, and reference the target role in the summary. An auto-pick mode ranks your whole resume library by keyword overlap and lets the LLM choose the best fit.
A refinement loop lets you read the suggestions, add any real-world detail the resume left out, and re-tailor with that context woven into the right bullets. Results export to a server-rendered .docx that opens straight in Word. A folder-of-Markdown project library can be injected into the Projects section when relevant.
Resume tailoring Demo