Completely free AI resume tailoring for students and the job-seeking community. Paste any job description, get a match score, gap analysis, and an ATS-friendly PDF in under a minute.
No reviews yetBe the first to leave a review for Resunova
Maker
📌
Hey Product Hunt! 👋
I built Resunova because every AI resume checker I tried had the same problem: the AI lies to look smart.
It told me to "add metrics" to a résumé that already had nine quantified bullets. It gave me "improved" rewrites that were word-for-word identical to my originals. It listed "no tables detected" a good thing -> as a warning.
So the core of Resunova isn't the scoring model. It's the honesty pipeline that runs after it. Every claim the AI makes gets checked against your actual résumé text:
• Issues that contradict the evidence → dropped
• "Improved" rewrites identical to your original → never surface
• Rewrites that would delete your numbers or the companies you worked for → rejected
• And if the AI gets caught exaggerating more than once, we stop trusting its overall score entirely and recompute it from the category evidence
When we don't know a metric, you get a [X%] placeholder, we never invent numbers for your résumé. Your interview is not the place to discover your resume lied for you.
What you get in under 60 seconds, free, no account needed:
⭐ An 8-dimension score where every number under 95 explains itself
✍️ Bullet-by-bullet rewrites that keep your facts
📄 A tailored, ATS-safe PDF that matches the preview exactly
💼 160k+ live US jobs pulled daily from 8,000+ company career pages, ranked against your résumé with salary data, H-1B sponsor history, and real HR contacts
🎤 Interview prep generated from YOUR résumé : behavioral questions that cite your own bullets, a STAR story bank, and coding questions only when the role actually calls for them
It's completely free for students and the community.
I'd love the harshest feedback you've got. And here's a standing offer: upload a résumé, and if the AI says something dishonest that slipped past the validators, call it out in the comments, I'll treat it as a bug report and ship a fix today.
Report
Sharp! Since you don't require an account to use the tool, how are you handling data privacy and retention? Are user resumes stored temporarily or is everything processed strictly in-session and wiped immediately after?
Report
Maker
@hasti_leo Thank you for using our tool. Yes, we are using FERPA guidelines to maintain data privacy across the board. Users data is stored as long as the user intends to keep the account.
How does the match score actually work under the hood, like is it just keyword matching or does it understand context and transferable skills too?
Report
@acsefa73688 Good question, and the honest answer is: it's a hybrid, and the split matters.
When we ingest a posting, an LLM reads the full job description and extracts structured requirements: each one gets a canonical name, a type, and an importance bucket (must-have vs nice-to-have). That's where context lives, the extractor understands that "experience building consumer-scale event pipelines" is a streaming-systems requirement even though the JD never says "Kafka". Hope I answered your question.
Sharp! Since you don't require an account to use the tool, how are you handling data privacy and retention? Are user resumes stored temporarily or is everything processed strictly in-session and wiped immediately after?
@hasti_leo Thank you for using our tool. Yes, we are using FERPA guidelines to maintain data privacy across the board. Users data is stored as long as the user intends to keep the account.
@parth_bhodia Smart move, Parth! I like the idea
How does the match score actually work under the hood, like is it just keyword matching or does it understand context and transferable skills too?
@acsefa73688 Good question, and the honest answer is: it's a hybrid, and the split matters.
When we ingest a posting, an LLM reads the full job description and extracts structured requirements: each one gets a canonical name, a type, and an importance bucket (must-have vs nice-to-have). That's where context lives, the extractor understands that "experience building consumer-scale event pipelines" is a streaming-systems requirement even though the JD never says "Kafka". Hope I answered your question.