The ones your own lines invite. A recruiter reads the resume in front of them and asks about the claim that is least defended — a number with no method behind it, a verb that hides how many people were involved, a date range with a hole in it. Upload the resume and you get those questions back, each pinned to the line that produced it.
Check my resume freeFree: full analysis plus your 3 most likely questions. $19 once for all 10 with STAR answers. No account, no subscription.
Most "top 50 interview questions" lists are indexed by role: backend engineer, product manager, marketer. That is the wrong index. Two backend engineers with the same title walk into the same room and get asked entirely different things, because one of them wrote "reduced p99 latency from 900ms to 250ms" and the other wrote "improved system performance". The first invites a specific, answerable question about method. The second invites a hostile one about whether anything happened at all.
This is why generic question lists feel useless in the room. You prepared for the average candidate's interview. The interviewer is holding your document.
A number with no method. "Cut costs by 30%" is an invitation to ask how it was measured, over what baseline, and what got worse. If you cannot answer the baseline question the number reads as decoration.
An ownership verb with no scale. "Led the migration" — of how many services, with how many engineers, over how long, and what did you personally do that nobody else could have. Interviewers probe "led" more than any other word on a resume, because it is the cheapest one to write.
A date range with a hole in it. Gaps are found on the resume, not in the conversation. The question is already formed before you sit down; the only variable is whether you have an answer.
A skill listed but never used in a bullet. Kafka in the skills row and nowhere in the experience section is the single most reliable place a technical screen starts, precisely because it looks padded.
Free gives you the whole analysis of the resume plus the three questions you are most likely to be asked, each one attached to the exact line that triggers it, and a sample rewrite of one line so you can see the standard. That is enough to know whether the read is accurate about your own document — which is the only way to judge a tool like this.
The $19 Interview Prep Pack, bought once and not by subscription, adds the full rewrite as PDF and Word, all ten most-likely questions ranked by probability with STAR answers built from your real experience rather than invented, a 60-second walk-through script for "tell me about yourself", a red-flag playbook for gaps, layoffs and title inflation, and a defensibility audit against the specific role you named.
Five worked examples are published in full, so you can read the analysis before handing over anything of your own: a backend engineer in fintech, a B2B SaaS product manager, a data analyst moving to data science, a frontend engineer at a startup and a DTC growth marketer.
This predicts the questions your document invites; it cannot know what your particular interviewer cares about, and a panel that has decided to talk only about system design will do that regardless. It will not invent achievements — if a bullet has no substance underneath it, the output says so rather than inflating it, which is not always what people want to hear. It is not an ATS keyword scorer, and it does not claim a score. The parsing happens in your browser, but the text does go to an API to be analysed, so this is not a fully local tool; the resume is held for up to 24 hours and then deleted. Analysis quality is strongest in English.
Resbetter · STAR answers from your own resume · Tell me about yourself · Explaining an employment gap · Resume checker with no signup · Privacy