Hey everyone,

I built a free AI-powered resume reviewer to help students and job seekers improve their resumes before applying.

It provides:

1.ATS compatibility analysis

2.Section-by-section feedback

3.Actionable improvement suggestions

4.Interactive report

5.No sign-up required

I'm looking for honest feedback from people who are actively applying for jobs.

Specifically:

Was the feedback actually useful?

Was anything confusing?

What feature would make you use it again?

Did you notice any bugs or UX issues?

🌐 https://macoostudy.info⁠�

I'd really appreciate any honest feedback—positive or negative. I'm actively improving it based on user suggestions. Thanks! 🙌

24 F on F-1 OPT. I know I should not lose hope and I won’t. But it’s so frustrating. I have a bachelors degree in accounting and auditing and a masters degree in finance. I also have about a year of experience in risk advisory working at big 4 at my home country and full 10-month internal audit internship experience at a fortune 500 in US. I’m targeting internal audit roles that require 0-5 years of experience. But only get rejections, barely get the HR phone screen, and when I do, it doesn’t go forward with the reasons either they don’t want people who doesn’t have a citizenship/GC or they want more experienced/more specialised experience. I’ve been preparing for my CIA part-1 exam and planning to take the exam this month, in hope that helps a bit. I’m really tired of not understanding what to do. The roles that I think i can absolutely nail are internship for which i get auto filtered since i’m not in school anymore. The entry level (0-2/3) years of experience roles are also rejecting me god knows why (am i under qualified/overqualified).

I also tried a different angle where I target Financial Analyst/FP&A full-time roles and internships, because I’m interested in that as well and because I saw some of my fellow finance graduates with 0 experience land financial and senior financial analyst roles, but I’m getting no luck there as well.

I have tailored my resumes to the jobs i apply, nothing is working. I sometimes even say no to the sponsorship question, just to test whether i’ll get an hr phone screen, and sometimes i do, but nothing goes forward. The only reason/feedback i get is “they have found other candidates that better align with the role” when i know i am very well perfectly aligned with the JD for some roles that I applied, the exact work i have done.

And I do not wish to spend the a huge amount on the consultancies who random stuff your resume with keywords and apply and apply, and when you get a job, they take a huge percentage of your first years salary. And still don’t promise you a job.

I would like any suggestions if you guys have.

I work in HR and open these files every week, so a few honest things that cost good people interviews:

1. The software has to find you first. Most mid-to-large employers use an ATS — it parses your resume into a database, and recruiters search it by keyword. If the posting says "client relationship management" and you wrote "managing customers," you don't show up. Mirror the posting's exact words.

2. Duties ≠ achievements. "Responsible for managing the team" tells me nothing. "Led a team of 12, cut turnaround 30%" tells me you make things better. Verb + what you did + a number.

3. One column, standard headings, no graphics. Fancy two-column/text-box designs confuse the parser and your work history gets scrambled. Clean and boring beats pretty and broken.

4. You get a ~6-second human scan. Name clear, a 2-line summary up top, recent role obvious, white space. Crowded page = crowded thinker.

Happy to answer resume questions in the comments — drop yours.

Hlo guys any platform (free) suggestions for tailoring my resume with ai after uploading the jd ?

I’ve spent the last year building AI infrastructure that interfaces directly with recruiting platforms. In the process of figuring out how to automate applications without breaking the parsers, I had to learn exactly how these systems digest PDF files.

If you're wondering why your Easy Apply goes into a black hole, or why Workday makes you retype everything, here is what’s actually happening under the hood.

1. Workday (The Keyword Filter) Workday is incredibly keyword-density focused. When it parses your resume, it actively scores it against the job description and produces a match percentage. Recruiters often set filters based on this score. If you’re below a certain threshold (often 60-70%), no human ever sees your file. This is why tailoring your CV for Workday actually yields callbacks, but the manual data entry makes it a nightmare to do at scale.

2. Greenhouse (The Structure Stickler) Greenhouse parses section headers very aggressively. If your resume gets creative and uses "My Professional Journey" instead of "Experience," or "What I Know" instead of "Skills," its parser often drops those fields entirely. Fancy dual-column Canva layouts routinely break Greenhouse's text extraction.

3. Lever (The Data Validator) Lever is much more forgiving on crazy formatting and columns, but it is brutally strict on contact info extraction. If your phone number isn't in a standard international format (e.g., +1, +33), it might map it to null.

4. Ashby (The Modern Parser) Ashby is newer and has a significantly better parsing engine than the legacy systems. It handles multi-column and heavily designed resumes better than most. However, if you embed actual tables inside your PDF to align your dates, it will still scramble the text sequence.

5. SmartRecruiters (The Aggressive De-Duplicator) SmartRecruiters has solid text parsing, but it is incredibly aggressive on candidate de-duplication. If you apply to a new role but you previously applied there 3 years ago with a different email, it will often merge your profiles based on name/phone number and sometimes prioritize your old resume data in the recruiter's primary view.

TLDR on how to beat them:

1. Stop using columns. Plain, single-column formatting is the only way to ensure Workday and Greenhouse read your bullet points in the correct order.
2. Use standard headers. (Experience, Education, Skills). Do not deviate.
3. Mirror the exact phrases. If the JD says "Python3", don't write "Python". Workday's exact-match logic is not always smart enough to bridge the gap.
4. Always use PDF. DOCX parsing remains highly inconsistent across older ATS versions.
5. Standardize your phone number. Use the +[country code] format.

I'm deep in the weeds on this stuff right now. Let me know if you have questions about specific ATS systems or how different AI autofill tools interact with them!

let me preface this by saying i am not a career coach. i'm just a guy who's been unemployed for 16 months, has sent somewhere north of 400 applications, and accidentally ran an experiment on himself.

around month 8 i hit a wall. not the emotional wall, i'd already hit that one three times. this was the practical wall where i genuinely could not spend another two hours rewriting my resume for a role that would ghost me in four days. so i stopped. i built one clean generic resume and just blasted it. for about six weeks i probably sent 60 something applications with that same file.

total callbacks that month: one. and i'm pretty sure that one was a mistake because the recruiter opened with 'hi is this jennifer."

so i went back to tailoring. but this time i actually paid attention to what i was changing and why, instead of just rewriting random lines hoping something would stick. and the next month i got six callbacks. same market, same experience level, same types of roles. six versus one.

the thing i figured out is that there's tailoring that exhausts you for no reason, and there's tailoring that actually does something. i was doing the first kind for months. rewriting whole paragraphs, reworking my whole summary, moving sections around. none of that is what matters. what matters is much more boring and mechanical and honestly once i understood why it worked it took me about 15 minutes per application instead of two hours.

the short version is that the system doesn't read your resume the way a person does. it runs a search. and your resume either contains the words that search is looking for or it doesn't. so tailoring isn't about sounding good, it's about matching. and the posting hands you the exact words you need to match, for free, every single time.

i know that probably raises more questions than it answers and i'm happy to get into the specifics, the what i actually change, the what i don't touch, the how i read a posting to find the keywords that matter versus the filler. i can also talk about what i do when i'm applying at volume and don't have 15 minutes for every single one, because there's a version of this that gets faster the more you do it.

also want to be realistic about the part that still suck. even doing this right, even when i know my resume matched and i know it was seen, i still get ghosted constantly. the callback rate going up doesn't make the silence easier. it just means the silence is no longer about the resume, and weirdly that's both better and worse. better because at least i know what's working. worse because now i have to figure out what's happening in the interviews.

anyway. if you're in tech and you're in the grind right now, what are you struggling with most? the getting seen part or the getting past the first call part? because the fixes are pretty different and i've been burned by both.

I spent a stupid amount of time figuring out why my resume was invisible. I talked to maybe a dozen recruiters over coffee, on calls, in DMs. They all told me versions of the same thing. The machine reads your resume before the human does. And the machine is dumb.

Here is what actually happens when you click submit

The five steps where things break

Your file goes through a pipeline. Ingestion, extraction, tokenization, mapping, storage. Sounds fancy. It's not.

Ingestion just checks if you're a PDF or DOCX and if you're under the file size limit. Boring.

Extraction is where it gets messy. The system pulls the text layer from your PDF. Not what you see. What the machine reads. These are different things. I learned this when a recruiter showed me my own resume in their system. Half my bullet points were gibberish. My beautiful custom font rendered perfectly on screen but encoded so badly the parser choked on it.

Tokenization breaks your text into searchable chunks. "Cross-functional" might become "cross" and "functional." Or "crossfunctional." Depends on the parser. A recruiter searching "cross-functional" might not find you if your resume got split weird.

Mapping is the worst step. The system guesses what's your name, what's your job title, what's a date. It gets this wrong constantly. I saw a resume where "Senior Product Manager | Google" parsed as title="Senior" and company="Product Manager." That person never showed up in any search for Product Manager.

Storage is just dumping everything into a database. If the previous steps broke, you're in there but unfindable.

Test your resume right now

Open your PDF in Chrome. Ctrl+A, Ctrl+C. Paste into Notepad. What you see is exactly what the machine sees.

If your skills section is blank? The parser couldn't read it. If your dates are missing? The parser stripped them. If sections ran together into one blob? Your layout broke the reading order.

This test takes 30 seconds. Most people never do it.

Encoding failures that kill you

Smart quotes from Word. Those curly " and ' characters. They often become or ™ or just vanish. Use straight quotes. Always.

Em-dashes and en-dashes. Copy-pasted from job posts usually. They disappear or split words. Replace with hyphens.

Fancy bullets. Arrows, checkmarks, diamonds. They become ? or blank space. Use hyphens or asterisks.

Accented characters. José, François. Sometimes work, sometimes become "Jos" depending on which ATS version the company bought. I saw this break search at a major provider.

Tables and columns. Multi-column layouts read left to right across both columns, line by line. Your skills on the left and experience on the right become "Python Senior Engineer SQL Google." Nonsense.

Headers and footers. Some parsers strip them entirely. Others dump them into random body text. Never put your contact info there.

The PDF vs DOCX thing

I tested both. A lot.

PDF keeps your formatting locked. Looks consistent everywhere. But text extraction fails more often, especially with design tools.

DOCX parses cleaner because there's no extraction layer. The system reads it natively. But formatting shifts between Word versions. And some systems flag macros.

My numbers: PDFs got 15% more callbacks for design-heavy roles. DOCX got 8% more for traditional text-heavy formats. Default to PDF unless they specifically ask for DOCX.

The hidden confidence score

A recruiter told me about this. Most ATS systems assign a confidence score to parsed resumes. Low confidence means you go to the bottom of the pile. Or a manual review queue that nobody checks.

What tanks your score? Weird section headers like "My Journey" instead of "Experience." Missing expected fields. Extraction errors. Mixed date formats. Inconsistent bullet styles.

High confidence means you surface first. Be boring. Be predictable. Be parseable.

The semantic search lie

Some ATS vendors sell "AI-powered semantic search." It understands concepts, not just keywords.

I tested it. Uploaded a resume with "data visualization." Searched for "data storytelling." No match. Searched "Python" against "PySpark." No match. Searched "project management" against "PMO." No match.

The AI is marketing. Recruiters use boolean keyword search because it's reliable. The system finds what they type. Not what they mean.

What I actually do now

I rebuilt my resume for mechanical readability. Standard headers. Consistent dates. Hyphen bullets. Single column. Arial or Calibri. Saved from Word, not Google Docs. I run the Notepad test on every version.

This is tedious. I spent months obsessing over character encoding and tokenization patterns. I had dreams about PDF text layers. The burnout was real.

I eventually started using tools that handle the mechanical optimization. CVnomist, Resumaniac, and Claude for specific tasks. They pull keywords from job posts, match them to my experience, and keep the output parseable. I review and send. Five minutes.

Don't use generic ChatGPT for this. It adds fancy formatting that breaks extraction. Smart quotes. Creative headers. It doesn't know about text layers or encoding. The specialized tools do.

Your checklist

Test your PDF. Ctrl+A, Ctrl+C, paste to Notepad. Fix anything broken.

Use standard section headers. "Experience" not "My Journey."

Use one date format. Never mix them.

Use simple bullets. Hyphens only.

Use standard fonts. Nothing custom.

Put critical info in the body, not headers or footers.

If all of this is clean, you are findable. Then you can worry about keywords and titles and all the other stuff.

The machine is dumb, make it easy for it.

I’m curious about the recruiter side of ATS systems like Workday, Greenhouse, Lever, and Ashby.
When a recruiter opens a job with hundreds of applicants, what is the default view most recruiters see?
• Are candidates shown in the order they applied (oldest first)?
• Most recent applicants first?
• ATS match score/ranking?
• Referrals first?
• Candidates with recruiter activity first?
• Something else?
For anyone who has used these systems as a recruiter or hiring manager:
1. What is the default sorting in each ATS?
2. Can recruiters easily change the sort order?
3. In practice, how do recruiters usually review applicants?
4. Does applying early actually help, or do strong matches get surfaced regardless of application date?
Would love insights from recruiters who regularly use Workday, Greenhouse, Lever, or Ashby. Thanks!

Hello Everyone !! this is my first reddit post. Internship season is approaching and like others I am also making resume (first time) for SDE Role . Currently I have achieved ATS score 77 and want it to improve to 90+ score.

key things I added in my resume:

1. Education (college name, 10 th and 12th school names and grades achieved) .
2. Experience (I wrote undergraduate student researcher bcz I did research work under the supervision of a proffessor)
3. Projects (mentioned 3 projects telling what they are, project's metrics and its impact)
4. courses (relevant) taken in college (did not write grades achieved in them)
5. skills
6. Leadership (in college 's society/clubs .... For example- core member of xyz club . NOTE : I did not expand it)

Please give tips to improve the score.
Thanks