Agentic Coding IS NOT Software Engineering

Agentic coding isn't software engineering. If you've been coding for a while, it doesn't give you the same rewards. Yet some developers are all too eager to let their skills atrophy and outsource the fun parts of being a developer to an AI, instead of fixing the root issue - their skill issue. This is mainly targeted towards junior developers, but a lot of this applies to more experienced developers as well. Or it will, once they let their skills atrophy.

---------------------------------------------------

💻 Learn Frontend (20% off): https://scrimba.com/the-frontend-developer-career-path-c0j?via=donthedeveloper
⚙️ Learn Backend (25% off): https://boot.dev/?promo=DONTHEDEVELOPER
🤖 Learn AI Engineering (20% off): https://scrimba.com/the-ai-engineer-path-c02v?via=donthedeveloper
🐱 Learn NestJS (free) - https://scrimba.com/nestjs-c0n7djgjma?via=donthedeveloper
🧠 Advanced Coding (40% off) - https://app.codecrafters.io/join?via=donthedeveloper

👥 1 on 1 Mentorship - https://calendly.com/donthedeveloper/coaching
👁️‍🗨️ Join Discord - https://discord.gg/TpQe2k8Ab3
🐦 Follow on X - https://x.com/thedonofcode
👾 Follow on Twitch - https://twitch.tv/donthedeveloper

Disclaimer: Some product links are affiliate links which means if you buy something I'll receive a small commission at no extra cost to you.

Episode Transcript

Don Hansen (00:00):
If you enjoy agentic coding more than you do coding
without AI, you don't reallylike coding.
Stop telling people you likecoding.
Of course, you don't likecoding.
Not nearly as much as youthought you did.
Agentic coding is so I put moretime into really trying to

(00:22):
utilize AI to build out certainfeatures.
I have apps that would greatlybenefit me to build out with AI,
and I just hate every bit ofit.
So what is agentic coding?
What does that mean to adeveloper?
That means several things.

(00:42):
Um, one, I don't know if you'velike some developers just work
in a very small shop and theydon't really do code reviews.
They don't get code reviews.
So maybe you're not used tothis.
But if you are working with ateam that does code reviews, I
don't know about you, but Iwould rather be building
features than reviewing otherpeople's features that they

(01:04):
built.
I don't like code reviews.
I do them to help others, butit's not my favorite part of
being a developer.
It's one of those parts whereit's like, you just gotta do it.
Um, I don't really like settingup the feature in local and
testing it out.
I don't care.
I like building my ownfeatures.

(01:24):
I like building stuff.
You are a code reviewer withagenda coding.
You are constantly reviewingthe AI's shitty ass code.
Unless you keep it very localscope and do it little by little
by little, um, have really goodspec files, and we'll talk
about that in a second, but youare a code reviewer.
That's all you are.

(01:45):
You're not a developer.
There's so much more to being adeveloper than just code
reviews.
And if you honestly just stickto code reviews with agenda
coding and having it build outfeatures for months, you're
gonna get extremely rusty.
You're not going to be a goodsoftware engineer.
A lot of people that areleaning on these AI tools
extremely heavily as a softwareengineer are going to atrophy

(02:07):
with skills that are stillrelevant.
And you really need to think toyourself: if all you like doing
is code reviews because youlove agenda coding, why are you
trying to become a softwareengineer?
It's okay to be a little bittech savvy and kind of be, you
know, go towards a projectmanagement route, indie hacker

(02:28):
route, and just get stuff outquickly.
Well, okay is maybe not theright word.
It is a path that you canchoose.
Why are you trying to become asoftware engineer?
There's so much depth tobecoming a software engineer to
be competitive right now.
There are other options.

(02:50):
I'm being serious about this.
If you truly love agendacoding, you might be going down
the wrong path of trying tobecome a software engineer.
So another thing to make agendacoding very effective is
creating really good spec files.
Um, product managers can dothis very well, especially tech
savvy, technical projectmanagers, they can do this very

(03:11):
well.
They're used to building out alot of issues and they like that
sort of they they like thespecs, they like that aspect of
project management where theyreally have to think through and
outline all the pieces of thefeatures, all the pieces of the

(03:32):
solution, because they've verythoroughly identified the
problem that they're trying tosolve, right?
If you need to build out a specfile, you're probably going to
start with something a littlebit more generic, but you have
to understand the problem thatyou're trying to solve, even
think about how the solution isbroken down.
When you deliver this to AI, itcan be much more effective in

(03:56):
generating code.
But you're a project manager.
I don't mind creating tickets,I don't mind breaking things
down.
I actually find that helpfulsometimes, but it is an aspect
of being a software engineer.
And sometimes something youdon't really have to deal with,

(04:17):
especially kind of likejunior-ish level.
Um, but sometimes you have veryheavy project management where
they are doing a lot of this foryou.
I don't think I would thrive insomething like that.
I like kind of being, I likebeing close to the issues that
I'm trying to solve.
I like really understanding theuser's complaints and their

(04:39):
frustrations.
But you want to be really goodwith agenda coding, you need to
be a good project manager andyou need to love code reviews.
Those are the two main things.
Both of those, to spend most ofmy time on both of those,
sounds horrible.
That sounds so fucking boring.

(05:02):
I don't know how anyone thatsays they love software
engineering that wants to be asoftware engineer would settle
for that.
It's like the worst parts ofbeing a software engineer, like
some of the best parts are justsolving really, really hard
challenges.
If you're using AI to try todebug your code, what are you
doing?
That's fun.

(05:22):
Why aren't you having fun withthat?
You're going to grow and getfaster eventually, but you are
just outsourcing the debuggingpart of it, which is very often
lackluster.
But, you know, if it's scopelocally, it doesn't require a
ton of context.
It doesn't require a lot ofdifferent pieces of your code

(05:43):
base.
It can help, it can solve somebugs pretty quickly, but don't
you enjoy doing that?
It's not an excuse that AI doesit faster.
If AI does it faster, it meansthat's a skill issue.
AI is not that good, especiallywith a lot of context.
It's not that great right now.

(06:04):
And if it's faster than you,that should be a signal that
that's a skill issue that youcan improve and get better.
Don't you want to get better?
Because if you don't, you areexactly the type of developer
that will be replaced by AI.
You just are.
I don't understand how peoplehave they just so easily adopt

(06:28):
this mindset of accepting such alow skill level for yourself.
To me, that's such a patheticthing to accept.
Why don't you want to getbetter?
Why do you want to outsourceyour intelligence, your
problem-solving ability, yourability to code to something

(06:51):
else?
What you're just handing awayyour job.
Like I it just doesn't makesense to me when software
engineers push agentic coding.
If you're an indie hacker,fine.
If you're a project managerthat is tired of working with
software engineers, fine.
Like I I get those.
I guess this is it's just arant.

(07:13):
I here's what I love aboutsoftware engineering.
And I think it aligns with alot of other software engineers,
at least that I've worked with.
I love organizing a code basethat grows.

(07:34):
I like trying to make itscalable.
I like making trade-offs withimplementations and owning that.
I like facing the demons thatI've created with one of my
implementations, getting pissedoff at myself or doing something

(07:56):
so dumb, and then owning thatand getting better from it.
I don't care about correctingAI.
I'm not helping anyone by doingthat.
I feel like it's such a wasteof time.
I don't mind code reviews fordevelopers that need to grow.
And I mean, code reviews fordevelopers that are better than

(08:18):
me, it's still helpful.
More eyes on something is stillhelpful.
A different perspective isstill helpful, but we're all
kind of just helping each othergrow and build better stuff.
I don't like mentoring an AI.
It's not human.
I don't give a shit if it getsbetter.
I don't care about AI.

(08:38):
I don't care about it growing.
I like mentoring.
I can't mentor an AI.
That's not mentorship.
Like when you mentor juniordevelopers, there's fulfillment
in that, there's rewards inthat.
Like I feel rewarded for seeinga junior developer grow and
being a big reason of why thathappens.

(08:59):
How can you care aboutinvesting so much time into
trying to improve your AI?
Why don't you just improveanother junior developer?
I just it it it feels so weirdwith the direction that we're
going with this blind acceptanceof something that isn't human,

(09:24):
right?
And that's not even that good.
Like if it would replacesoftware engineers, which it I
it was supposed to, what, twoyears ago?
I I don't even know at thispoint.
Fine, software engineering canevolve, but LLMs after the
release just continue to getworse and worse and worse and

(09:45):
worse.
I don't like fixing all thebugs that they create in my code
base.
It's a pain in the ass.
How is this a pleasurableexperience for people?
You want to know why?
It's because the people thatare pushing agentic coding are
either extremely pressured bymanagement, but likely have a
skill issue that they don't wantto improve.

(10:07):
They're just desperate to pushout something fast.
It doesn't matter that it'sscalable or good code, but it
works.
And there are product managers,there are executives at
companies that that is all theycare about.
And as software engineers,we've had a fight against that
for a very long time.
Because we have to continue toclean up that fucking garbage
over and over and over.
But if you want that in yourcode base because you feel

(10:30):
pressure, fine.
But if you step outside ofthat, because I think that's
kind of niche territory, atleast at the moment, when you
step outside of that and you'rebuilding something with a team
and you're a good developer, youyou realize how unmaintainable
this becomes.
When you are leaning heavily onagenda coding, you realize that

(10:52):
it is just producing a veryunscalable application, very
unmaintainable application, andthat snowballs over time, and
your experience gets worse andworse and worse.
The people that are pushing foragenda coding, and I think this
is really important, are peoplethat don't care to be really

(11:13):
good developers.
They just want to push out aproduct quickly.
That's it.
They don't really enjoy codingas much as you think they do.
They don't like being asoftware engineer as much as you
think they do, but they mightbe loud, they might have a lot
of followers.
But these types of people,they're product managers.
They just see coding as a meansto an end.

(11:37):
And I've seen this in a lot ofmy comments.
It's like coding is a means toan end.
I've seen that comment over andover and over.
I know exactly the type ofdeveloper, they're not a
developer.
I know exactly, but they claimthey are.
I know exactly the type ofperson that says that.
I've met a lot of those people.
They aren't going to dive deep.
They're not trying to become areally good software engineer,

(11:59):
despite what they tell you, butit's just noise on the internet
right now.
So if you truly enjoy coding,if you enjoy the craft of it, if
you enjoy the creativity of it,if you like architecture, if
you like seeing your shitty codeimprove over time, if you like
helping other developers andseeing their code improve over

(12:21):
time, and you just respectyourself enough to want to
become a good software engineerin this field, you want to lift
others up.
I promise you, you are notalone.
There's just so much noise thatis completely misrepresenting

(12:43):
what software engineering is.
It's misrepresenting who reallygreat software engineers are
and how they got there.
I promise you, agentic codingis not the way, and you are only
going to shoot yourself in thefoot if you are actually serious
about software engineering.
Please, please put down the AIand fix your skill issues.