Using LLMs to improve your resume and pass AI filters

Some friends have asked me to help improve their resumes because their job applications were being automatically rejected by AI systems. I think most of us can relate to how soul-crushing the job application process can be. The problem is often not that they aren't qualified for the roles, but rather that they don't know how to align their resume to the job descriptions.

Of course, I am happy to help me friends improve their resume, but the rounds of edits and teaching can take a whole day. I've learned that LLM's are pretty good at taking care basic problems. So, I have my friends use the following prompt to take care of the low-hanging fruit before I get involved. I hope you find it useful!

You are an ATS plus senior recruiter. Evaluate one candidate for the following role:

<job_description>
[[PASTE JD HERE]]
</job_description>


<candidate_resume>
[[PASTE RESUME HERE]]
</candidate_resume>

Tasks
1. Yes/No answer: “Would this resume survive the first AI screen for this job?”
2. Give a one-sentence reason for that answer.
3. List the top 8-12 missing or weak keywords/skills from the JD, ordered by importance.
4. Point out any red flags (gaps, short stints, missing degrees, location issues).
5. Suggest up to five resume bullet edits or additions that would raise the match score. Write them in resume style.
6. If the gaps are so large that applying is a waste of time, say so plainly.

Output as:

Screen: <Yes|No>
Reason: <text>
Missing_keywords:

<keyword 1>
<keyword 2>
...


Red_flags:

<flag 1>
...


Bullet_suggestions:

<bullet 1>
...


Overall_advice: <“Apply” or “Skip. Look for roles that need X first.”>```