Hiring Manager's AI Resume Rules
Key Points
- Do not let a large language model write your entire résumé, because its default “house style” will make you sound generic and blend in with other applicants.
- Avoid using an LLM to answer interview or application questions, as the responses tend to be vague, word‑y, and fail to showcase the clear, incisive thinking recruiters look for.
- Refrain from using an LLM to “tune” or keyword‑pack your résumé for each job description, since this strips away your personal voice and often inserts keywords in the wrong places, making the document look mechanical.
- Only consider AI tools if you’re skilled at crafting system prompts that fundamentally change the model’s style to match your unique tone; otherwise, the AI’s inherent style will dominate and hurt your chances.
Full Transcript
# Hiring Manager's AI Resume Rules **Source:** [https://www.youtube.com/watch?v=iP8p7YjGG9k](https://www.youtube.com/watch?v=iP8p7YjGG9k) **Duration:** 00:14:36 ## Summary - Do not let a large language model write your entire résumé, because its default “house style” will make you sound generic and blend in with other applicants. - Avoid using an LLM to answer interview or application questions, as the responses tend to be vague, word‑y, and fail to showcase the clear, incisive thinking recruiters look for. - Refrain from using an LLM to “tune” or keyword‑pack your résumé for each job description, since this strips away your personal voice and often inserts keywords in the wrong places, making the document look mechanical. - Only consider AI tools if you’re skilled at crafting system prompts that fundamentally change the model’s style to match your unique tone; otherwise, the AI’s inherent style will dominate and hurt your chances. ## Sections - [00:00:00](https://www.youtube.com/watch?v=iP8p7YjGG9k&t=0s) **AI Resume Writing: What to Avoid** - A hiring manager explains why applicants should not let large language models fully craft their resumes, warning that the AI's generic “house style” can erase individuality and make candidates sound indistinguishable. ## Full Transcript
we need to talk about resumés and AI
because there are a lot of people who
are giving really conflicting
perspective and I'm a hiring manager
I've been hiring in this cycle and I
have fairly strong opinions about where
AI can effectively be used by applicants
and where it should not be used so I
want to break that down into two lists
and I'm going to go into a fair bit of
granular detail that's what a nice long
form video like YouTube is about so
first a good list of things that you
should not be using a large language
model for during your job application
process are you ready number one do not
use a large language model to write your
whole resume for you don't do it the
reason why is that large language models
have a house style unless you are
reasonably a Adept an expert at
prompting large language models with a
system prompt that substantially changes
their house style so that it sounds very
very very much like you and it's
distinctive and it's individual then
don't do it and if all of that didn't
make sense to you then you're not using
a system prompt that is going to help
you and you should definitely not be
doing it fundamentally unless you have a
fair bit of work put in an llm is going
to sound like an llm Claude has a house
style Gemini has a house style
open Ai and chat GPT have a house
style do not let that house style into
your resume because so many other people
are doing so you end up sounding like
all the other applicants is that what
you want is that what you're looking to
do is to not stand out from the crowd
because that is typically what happens
when you let an llm write your resume
all right that's number one number two
if there are questions from the company
to you during the applicant process
don't use an llm to answer them same
reason if you use an llm you have a
house style and for for a second reason
those questions are typically designed
to measure incisive and careful thinking
and next token predictors which is what
a large language model is are not going
to give you the clarity of thought that
you need to stand out it's going to feel
like word salad maybe not inaccurate
word salad but word salad
anyway and if it feels even a little bit
fudgy or a little bit flabby or a little
bit sloppy then the person reading the
response is going to think that's the
way you communicate that you're not
incisive that you're not concise that
you're not clear and you don't want that
again it's not helping you stand
out okay third
thing do not use an
llm to tune your resume
I know it's really tempting when you
have a lot of applications to do to just
say to the llm please tune the resume to
match this job description but the
problem is you lose all of your own
style in that process and the house
style comes back and the llm isn't
necessarily inserting the keywords that
you want it to insert on the correct
bullets and if you are reading several
hundred resumés you start to get an ear
for when
people are using keyword packing and
llms tend to
be not super subtle about hiding the
keyword packing and that's actually one
of the things you have to do to write a
good resume yes you want to match the
keywords but it needs to feel
natural and in my experience an llm
written resume has trouble adjusting and
tuning and bringing those keywords in
without making it reasonably obvious
what's going on and it does stand out if
you've read a bunch of
resumés all right number
four don't use an llm for the Final
Phase of preparation for interviews and
I'll come back to this theme but at the
end of the day the reason why interview
preparation is so important is because
your brain needs to actually speak
something out to sound
natural so don't use it for interview
prep or you're going to sound like
you're reading from a large language
model output you're going to sound
really robotic and yes I know there are
some people who literally get the llm
answers printed onto a screen while
someone is interviewing them and it is
obvious and people do know and it's a
huge red flag because we need you to
think creatively and think incisively
and think for yourself and so if you are
practicing with an lolam you are
unlikely at the final stage of practice
to be able to articulate your answer in
your style using your words and your
voice in a way that feels natural so
that Final Phase of interview practice
has to be with another human has to be
with a mirror if you don't have another
human to practice
with okay and then the last one that I
would say that you don't do is do not
use a large language model that has
search capabilities such as chat GPT or
even
perplexity for company research that is
high fidelity so if you're doing
interviews and you get to like the
second round or Beyond you need to have
a really good idea of what the company
does so that they can imagine working
with you because you have so much
corporate knowledge right
already it is hard to get that level of
uh tactile fine grained detail if you
are using an llm as an interface and the
reason for that is that they are
designed to summarize and when they
summarize they do lose detail and you
need that detail to stand out from the
crowd so don't use them for that really
high fidelity research that you need to
do to truly stand out in subsequent
rounds of interviews okay that was a
list of five different things you should
not be using large language models for
but there are things you should be using
large language models for in the
application process and that's the
second list and that's what we're going
to get to now so what you should use
large language models for is one I think
they're great tools for rfit analysis if
you to think about does this role fit my
skills does this role fit my job
description is there a way I can
quantify that or measure that yes there
are pre-built tools like teal that do
that you can also prompt and work with a
large language model just in the chatbot
experience and get a great result just
by having them read the two documents a
job description print out and then your
resume and starting to have a
conversation about the fit and the
weaknesses and the gaps that's a super
useful way to use a large language model
to understand whether you fit a role
another useful way to understand an llm
this is number two get them to tell you
what is distinct about that role that
doesn't fit the other standard job
descriptions again you're asking the llm
to look through its enormous experience
I guarantee you it has read more job
descriptions in your job family than you
have ask it to look at an individual JD
and say what is distinct about this and
then press it and make sure that it's
really clear that's a great use for a
large language model because that will
help you to read and understand intent
because the large language model can go
beyond just saying this is a distinct
sentence it can tell you this is a
sentence and this is my supposition
based on my extensive Corpus of text the
whole internet of text on what the
intent behind that bullet was and it may
not be 100% right but it's right enough
that it will give you a clue that you
might not pick up on otherwise so I
think it's a good use just to like
understand what's distinct about resumés
and help you analyze them rapidly or
understand what's distinct about job
descriptions because those are the ones
you're looking
at so number
three it's really helpful to identify
missing
keywords and this is goes in the same
category of like comparing r fit and so
if if you're comparing r fit which is a
different thing you have to understand
at a high level if the if the rolles fit
that's number one number two understand
what's distinct about the role we just
talked about that number three if you're
going to apply do a second level of
analysis between your resume and the job
description and specifically look for
missing keywords and this is different
than keyword packing it's the first step
it's the analysis step where it calls
out the missing keywords llms do a great
job at that it may be on you to work in
the keywords but llms can help you pull
them out and understand what you have to
add in really
fast okay number four llms are really
good at helping you with blank page
syndrome so you know how I said earlier
in this uh in this YouTube that you
should not use a large language model
for writing a whole resume that's true
you should also not use it for those
questions I talked about that but if you
need a draft it's a great drafter it
gets you past staring at a blank page
and that can accelerate you because
there is nothing harder as a writer than
just looking at the blank page and
wondering where to start we just don't
have that problem which is a 10,000 year
old Problem by the way it's disappeared
in 2024 we don't have blank page
syndrome anymore or we don't have to
anybody can get off of the blank page
immediately and you should and you
should use a large language model for
that and then you should edit the heck
out of
it all right number five uh you should
be looking at how you can break through
fluency blocks in your overall
application process with a large
language model and I'm going to give you
two examples oftentimes people don't
understand networking and they and they
talk about hey how do I Network how do I
Network in my situation for my job role
how do I get past the
shyness large language models are not
going to give you personalized Insight
from someone who's done it before but if
you're getting past the blank page
syndrome for a skill like networking
they are super helpful to give you a
sense of what's in the box and
especially if you prompt them and ask
them to think about about it from a
strategic perspective and a
non-transactional perspective which is
how you should be networking they're
going to give you really thoughtful
initial advice like a first passet
advice if you're looking to understand
the skill in general it can be helpful
and almost everyone has gaps in their
skill set over the course of a job
application process there are areas
where you are strong maybe you're really
strong at resumé editing in areas where
you're weak maybe that's networking
maybe that's looking at other job sites
besides LinkedIn and so what you're weak
at is actually getting the full top of
funnel for job Discovery llms are that's
another thing they're good at like you
can talk to them and say I don't know
where else to go for jobs besides Linkin
can you help me find some that is a good
use for something like
perplexity basically you want to tune
the llm to give you a sense of how you
break through that skill Gap it's not
going to give you a high fidelity
excellent super professional final
answer on these things but it will get
you through that initial really um
painful feeling skill learning phase
faster and that's what you want it's
accelerating you and that's what you
should be looking
for okay and I also know at the top of
this YouTube video I talked about how
you should not use large language models
for the Final Phase of interview
practice and that is true you should not
be doing that but you should be using
large language models if you were
feeling stuck and need to practice your
initial interview structure and so if
you're doing the first half of interview
prep where you're determining do I have
the anecdotes do the anecdotes match the
anticipated questions do the anecdotes
have the right structure to them so that
someone who is a typical interviewer can
understand what I mean am I using the
star technique am I using the parade
technique am I opening myself up to
followup questions that are risky llms
are great as conversational tools to
help you get through that initial phase
of interview practice so absolutely use
them for that first half and then when
you actually need to feel natural and
practice it verbally that's when they're
not helpful anymore all right I'm going
to give you one final tip I believe this
is tip number seven for how to use large
language models in application prep and
sort of the whole job search process you
know how I said earlier in the video
that you should not be using a large
language model such as perplexity or
such as chat GPT with
for researching at a high level of
fidelity your company that you're
targeting great don't do it but you can
use a large language model to get a
highlevel view of the industry and the
competitors for the company because
again that plays to the system's
strengths it allows it to look across
the whole industry I guarantee you it's
read more articles than you have and to
synthesize it up and if you're looking
to synthesize and summarize a lot of
detailed information that you found that
maybe not
publicly available on the internet maybe
you maybe you've dug in the sec's
website and you found 10K filings for
these companies and they're recent so
they're outside the training data set
for the llm throw them in it's going to
be able to summarize them very very
rapidly and at a good enough degree of
fidelity that you can understand what
the competitor strategy is pretty
quickly and that is going to give you a
working knowledge of the industry and
competitors that makes you stand out
because typically you don't see someone
dinged in Loop Fe back sessions for not
having a high enough Fidelity view of
competitors what they're dinged for in
that area is not knowing competitors at
all or not knowing the industry at all
and large language models can get you
past that so there you go that is a
series of five ways not to not to use
large language models and seven ways to
use large language models as you're
going through your application and your
job search process AI is a powerful tool
for helping you just need to use it in a
way that supports your overall journey
and your ability to stand out and that's
what I wanted to call out here let me
know what I missed in the comments