Junior Developers in the Age of AI(thoughtfuleng.substack.com)
thoughtfuleng.substack.com
Junior Developers in the Age of AI
https://thoughtfuleng.substack.com/p/junior-developers-in-the-age-of-ai
22 comments
"There’s some scientific truth to this: 20-somethings are inherently narcissistic. Wisdom requires having a full frontal lobe."
This just became another boomer rant. Complete garbage.
This just became another boomer rant. Complete garbage.
I'm happy to hire juniors but they must still have relevant skills. One question I ask is "how does one build a coding agent" - if a junior is unable to outline the ideas around building and expanding coding agents, it indicates they are not aware of what industry they work in.
I hope you're hiring for an agent developer position, otherwise it's not going to be a very effective signal
Great LinkedIn post, thanks for sharing.
I guess I'm not aware of what industry I continue to work in.
I'm a senior dev and not sure I could answer that offhand.
More proof that a lot of idiots are interviewing.
Sorry but I don't understand why you ask this question, can you explain your train of thought?
How do we know it's "AI", and not just the fact that CS has gotten too popular and there's now a glut of new graduates?
https://cra.org/crn/2025/08/infographic-computing-bachelors-...
EDIT: it's also hard to account for the fact that job position titles aren't very consistent over time and I know quite a few new grads who got "senior" job positions
https://cra.org/crn/2025/08/infographic-computing-bachelors-...
EDIT: it's also hard to account for the fact that job position titles aren't very consistent over time and I know quite a few new grads who got "senior" job positions
Even before AI have noticed really small to mid size companies not having enough budget to hire junior developers that they know they'd have to train from the ground up. They'd rather take chances on senior developers and/or contractors.
I think there's a misconception that junior devs require all that much training compared to an experienced engineer. I rarely have all that many relevant skills when starting a new job. I have to learn their processes/tools/etc from reading the docs. So I'm probably coming in at like 20% better than junior.
Junior devs will have an easier time finding jobs in positions that operate in niche industries, because they offer onboarding for everyone.
Junior devs will have an easier time finding jobs in positions that operate in niche industries, because they offer onboarding for everyone.
If you’re only coming in at 20% above a complete junior you’re a slightly experienced junior elsewhere. You’re telling me your experience dealing with SDLC, incidents, abstractions, ubiquitous technologies like databases, CI/CD, k8s, etc only put you 20% ahead? Thats an absurd undersell.
> I have to learn their processes/tools/etc from reading the docs.
> So I'm probably coming in at like 20% better than junior.
There are firms that take that to heart, and there is indeed a lot of truth in it. A large amount of skills and knowledge just aren't transferable when switching jobs. But I think it's not hard to create more than 20% of the value. And even if it really is 20% of the value, the profit generated from the work might actually be more than the salary gap anyway, and the 1-year growth curve might be faster for a senior than a junior.
There are firms that take that to heart, and there is indeed a lot of truth in it. A large amount of skills and knowledge just aren't transferable when switching jobs. But I think it's not hard to create more than 20% of the value. And even if it really is 20% of the value, the profit generated from the work might actually be more than the salary gap anyway, and the 1-year growth curve might be faster for a senior than a junior.
That chart shown in the article is very interesting. But I think more important things happened in 2023 than just GPT3.5.
ZIRP ending, coding becoming a full blown meme ("day in the life of a wfh swe" videos on tiktok), and work from home exploding.
This combination of factors made it risky to make a mistake hiring an unqualified SWE. SWE salaries exploded AND money got more expensive. At the same time, the number of unqualified applicants exploded due to the EZ bootcamp wfh 6 fig meme, and on top of this, remote interviewing made it very easy to cheat your way into a job you were not qualified for. Since it's very difficult to gauge quality of SWE talent, this pushed companies very strongly in the direction of hiring senior talent. Of course, companies and managers do not want to announce that they do not know a good engineer from an awful one. So here comes a convenient excuse, the excuse that companies make for everything nowadays: "AI".
The article is correct though. This explosion in interest in CS, along with very powerful AI tools, has produced a very strong batch of junior engineers that are very very underpriced relative to their senior colleagues. If I were running a company right now and needed cheap, strong talent, there is never a better time than now. Recruit and hire exclusively junior engineers -- and only interview in-person.
ZIRP ending, coding becoming a full blown meme ("day in the life of a wfh swe" videos on tiktok), and work from home exploding.
This combination of factors made it risky to make a mistake hiring an unqualified SWE. SWE salaries exploded AND money got more expensive. At the same time, the number of unqualified applicants exploded due to the EZ bootcamp wfh 6 fig meme, and on top of this, remote interviewing made it very easy to cheat your way into a job you were not qualified for. Since it's very difficult to gauge quality of SWE talent, this pushed companies very strongly in the direction of hiring senior talent. Of course, companies and managers do not want to announce that they do not know a good engineer from an awful one. So here comes a convenient excuse, the excuse that companies make for everything nowadays: "AI".
The article is correct though. This explosion in interest in CS, along with very powerful AI tools, has produced a very strong batch of junior engineers that are very very underpriced relative to their senior colleagues. If I were running a company right now and needed cheap, strong talent, there is never a better time than now. Recruit and hire exclusively junior engineers -- and only interview in-person.
Software development compensation in the US id very much bi modal.
Most developers in the US work at banks, insurance companies, etc - “the enterprise” - in tier 2 cities. Those developers usually max out at around $160K-$175K or less and it hasn’t kept up with inflation. I did all of my “enterprise dev” up until 2020 in Atlanta. You can look at the compensation of well known Atlanta companies like Delta, Coke and Home Depot. You will see the same pattern in most other cities in the US outside of the west coast and NYC
Enterprise devs if they remain so will probably never crack $200K inflation adjusted. I don’t know any of my friends/former coworkers who are still software developers in their 40s who still live and work in Atlanta who are making over $160K. They are doing well enough though being married with two incomes.
The problem on that end is that it is easy enough to be a good enough generic CRUD LOB developer, it started being commoditized around 2015 and comp stagnated on the high end.
FWIW, I pivoted to cloud consulting specializing in app dev when a (full time with the standard 4 year comp structure) when a remote position at AWS ProServe fell into my lap (still work in consulting - not at AWS). We moved to state tax free Florida in 2022
On another note, it still doesn’t make sense to hire juniors. The difference in comp for a junior dev and senior dev on the enterprise /small SaaS startup side is at most 60K if you stay out of SFBay or NYC and set up shop in a second tier city. You probably won’t need the route. Look at the laughable comp being offered by most early YC startups. No statistically worthless illiquid equity in a private company is not what I consider “compensation”
Most developers in the US work at banks, insurance companies, etc - “the enterprise” - in tier 2 cities. Those developers usually max out at around $160K-$175K or less and it hasn’t kept up with inflation. I did all of my “enterprise dev” up until 2020 in Atlanta. You can look at the compensation of well known Atlanta companies like Delta, Coke and Home Depot. You will see the same pattern in most other cities in the US outside of the west coast and NYC
Enterprise devs if they remain so will probably never crack $200K inflation adjusted. I don’t know any of my friends/former coworkers who are still software developers in their 40s who still live and work in Atlanta who are making over $160K. They are doing well enough though being married with two incomes.
The problem on that end is that it is easy enough to be a good enough generic CRUD LOB developer, it started being commoditized around 2015 and comp stagnated on the high end.
FWIW, I pivoted to cloud consulting specializing in app dev when a (full time with the standard 4 year comp structure) when a remote position at AWS ProServe fell into my lap (still work in consulting - not at AWS). We moved to state tax free Florida in 2022
On another note, it still doesn’t make sense to hire juniors. The difference in comp for a junior dev and senior dev on the enterprise /small SaaS startup side is at most 60K if you stay out of SFBay or NYC and set up shop in a second tier city. You probably won’t need the route. Look at the laughable comp being offered by most early YC startups. No statistically worthless illiquid equity in a private company is not what I consider “compensation”
lil bro re-labeling himself as engineer instead of coder to keep job
> It’s why the best engineers try to write as little code as possible. They understand that each additional line is another thing that must be understood, remembered, and maintained.
As a long time SRE/DevOps, this is the key point that I believe people are overlooking in the "AI will let anyone write code!!"
Support does not scale linearly with additional software. If anything, it grows faster due to the "two systems == one connection, three == three connections, four == six connections" etc.
Sure, you could argue that LLMs could ALSO do the support but the reason you have people around to begin with is for when the technical systems break in weird and interesting ways. Similar to how planes have pilots even in the age of autopilots (the humans are, of course, still fallible).
Another argument might be: well, LLMs will write better software. My counter is that there weren't really excuses for not writing better software before but we still got buggy code with poor documentation. If anything, LLMs will make it easier to crank out barely maintainable and supportable code at even faster rates.
As a long time SRE/DevOps, this is the key point that I believe people are overlooking in the "AI will let anyone write code!!"
Support does not scale linearly with additional software. If anything, it grows faster due to the "two systems == one connection, three == three connections, four == six connections" etc.
Sure, you could argue that LLMs could ALSO do the support but the reason you have people around to begin with is for when the technical systems break in weird and interesting ways. Similar to how planes have pilots even in the age of autopilots (the humans are, of course, still fallible).
Another argument might be: well, LLMs will write better software. My counter is that there weren't really excuses for not writing better software before but we still got buggy code with poor documentation. If anything, LLMs will make it easier to crank out barely maintainable and supportable code at even faster rates.
Just post your prompt, I don't need to know what an LLM thinks of juniors in the age of AI and I don't know which parts you wrote.
It's just so transparent. A safe message using bland language. Nevermind the slop... the boring nothingburgers are even worse.
A few points here.
- Junior developers become senior relatively quick. Most senior developers get their senior title just a few short years into their careers. Typically in their late twenties or early thirties. Not judging but that seems to have been the state of affairs in software engineering for quite some time. I was a "senior" by age 30. I've learned a lot since then ;-).
- AI is increasing the demand for software because software just got cheaper to make so people want more, not less. The best software still gets made by senior people (with or without AI). That's why seniors are in demand. That should have an effect on their compensation. Which would be very welcome given the last few years of freelance rates dropping.
- Software creation is a team sport. Having a team mainly consisting of old and grumpy middle aged software engineers isn't necessarily the best thing (been there done that). It might be better to have a mix of young old and experienced/inexperienced. There's plenty of room for juniors in the industry.
- Juniors come in many forms. Some people have a lot to learn. Some people create valuable startups straight out of college.
- Software companies that are not constantly investing in their staff will have to deal with replacing their older staff as they move on, retire, or disappear for other reasons. A constant in flow of juniors that you can then train up and turn into valuable team members still works just like it always has.
- Juniors are a lot cheaper to hire than seniors. If there's a shortage of seniors, hire juniors instead and then train them up. That's what many companies were doing anyway. If anything a lot of silicon valley companies seem to have been preferring to hire juniors almost exclusively for quite some time.
- As long as the amount of programmers keeps on growing (and there are no signs of that slowing down that I know of), there will be more juniors than seniors. I'm not sure if the amount still doubles every five years but that seems to have been true for a long time.
- Junior developers become senior relatively quick. Most senior developers get their senior title just a few short years into their careers. Typically in their late twenties or early thirties. Not judging but that seems to have been the state of affairs in software engineering for quite some time. I was a "senior" by age 30. I've learned a lot since then ;-).
- AI is increasing the demand for software because software just got cheaper to make so people want more, not less. The best software still gets made by senior people (with or without AI). That's why seniors are in demand. That should have an effect on their compensation. Which would be very welcome given the last few years of freelance rates dropping.
- Software creation is a team sport. Having a team mainly consisting of old and grumpy middle aged software engineers isn't necessarily the best thing (been there done that). It might be better to have a mix of young old and experienced/inexperienced. There's plenty of room for juniors in the industry.
- Juniors come in many forms. Some people have a lot to learn. Some people create valuable startups straight out of college.
- Software companies that are not constantly investing in their staff will have to deal with replacing their older staff as they move on, retire, or disappear for other reasons. A constant in flow of juniors that you can then train up and turn into valuable team members still works just like it always has.
- Juniors are a lot cheaper to hire than seniors. If there's a shortage of seniors, hire juniors instead and then train them up. That's what many companies were doing anyway. If anything a lot of silicon valley companies seem to have been preferring to hire juniors almost exclusively for quite some time.
- As long as the amount of programmers keeps on growing (and there are no signs of that slowing down that I know of), there will be more juniors than seniors. I'm not sure if the amount still doubles every five years but that seems to have been true for a long time.
This article makes a number of the standard arguments for hiring juniors and a couple new ones:
- Companies should hire juniors [at their own perceived detriment] to improve the overall industry
- Hire juniors because strong companies are resilient to junior mistakes [rather than hiring seniors and also becoming resilient]
- Juniors learning fast inspires others
- Juniors will teach your seniors how to use AI
Perhaps you can see why most companies don't find these terribly compelling.
Personally, I suspect we're going to have to wait for capitalism to fix this. Senior engineers will age out and the supply will drop. This will increase the cost of seniors until low-cost juniors start to look like a better option to the median hiring manager.
- Companies should hire juniors [at their own perceived detriment] to improve the overall industry
- Hire juniors because strong companies are resilient to junior mistakes [rather than hiring seniors and also becoming resilient]
- Juniors learning fast inspires others
- Juniors will teach your seniors how to use AI
Perhaps you can see why most companies don't find these terribly compelling.
Personally, I suspect we're going to have to wait for capitalism to fix this. Senior engineers will age out and the supply will drop. This will increase the cost of seniors until low-cost juniors start to look like a better option to the median hiring manager.
We are in a transition phase and the standards to be employed are going up.
The best new grads and juniors will come to see AI as a “code generating machine” and “answer machine” and upskill themselves then come in at what we currently call mid level.