the developer job market is INSANE, 6 harsh realities

The software engineering job market, it's a massive cluster fuck right now. And honestly, sometimes I don't even feel like there is such thing as a job market. More like a bunch of creepy AI bots using AI generated resumes and applying to hundreds of fake AI jobs. It also seems like the only industry hiring right now, it's the Japanese porn industry. But judging by the average software engineer physic, I guess most of us will have no choice but stick to writing code.

And face this six harsh realities of the current job market. Software engineering reality number one, AI psychosis, it's reaching all-time highs. In the last couple of months, Silicon Valley has been force-feeding AI slop on the whole world. And AI use, it's now mandatory and enforced in tech companies like Amazon and of course, used as a performance metric. It seems like managers are pushing random tools on employees, even if most of them are half-baked and unhelpful.

So, many developers are using AI on fake tasks, it's also called token maxing. Just to basically avoid getting fired and singing along. And honestly, I don't blame them, I would probably do the same. It's kind of sad because this is what honestly, being a software engineer feels right now. You're being paid to basically make AI make yourself obsolete. And what used to be one of the best jobs in the world is now a freaking nightmare.

Meanwhile, most software products are actually getting worse. And I bet you've seen that, which is kind of crazy, right? Because if you think about the amount of marketing AI companies are putting out there, claiming AI coding will be the magic solution to all of our software problems. In 2026, GitHub already had 32.5 days of downtime. That's more than a month of some random GitHub service being down, which of course made projects like Ghostly, a fan of GitHub, actually live GitHub.

And what's even crazier is that this kind of AI obsession and AI push. It's happening just as these AI companies are increasing their token prices. GitHub co-pilot just published a huge price increase to use certain models. And in many cases, we're talking about a five-fold increase in token prices. So, a lot of software companies building software are developing dependencies on these AI companies, which the sooner they feel like they might have a a monopoly, they will jack up the prices. At this stage, honestly, AI coding will become so much more expensive than junior devs, which is actually good news. We need those juniors back in our development teams. They were actually cool people. Anyhow, AI psychoses also contaminated job postings, software engineering job postings, explicitly ask you for AI fluency, right? Things like experience with leveraging AI coding tools like Cloud code and cursor and codex or AI fluency, right, that you understand how LMs work and you probably built web applications with them. But what exactly does AI fluency even mean, right? Do you have to get a PhD in machine learning or do you just have to know how an LLM works or be able to to query an API? Well, I believe there are three levels of AI fluency, right? Number one, it's you are an AI user. You are mainly a user of LLMs, even if you are a software engineer. You use cloud code or open code or whatever fits you, right? And most of your knowledge, it's still about like how to build software in the traditional way, but using LLMs for velocity. So, you know the basics of harness engineering, you know, maybe context engineering and tokens and how to make the most off out of an LLM when you're building software. The next level is that you are what they call now an AI engineer, which is not a machine learning engineer, yeah? It's just a software engineer who can integrate AI into existing apps. For example, you can connect a frontend application to an LLM, and maybe you know about rag, about, you know, vector databases, what we use to do rag, etc. It doesn't mean you are able to build or optimize models, maybe deploy

maybe a basic deployment of a model, but you can integrate them in existing apps. And finally, there is the machine learning engineer. This is a mix between a software engineering and a machine learning engineer. They are able to, you know, fine tune LLMs, they can even design LLM architecture internal architecture. They can tailor LLMs to specific business domains. They have very strong foundations of data engineering and machine learning and a bit of linear algebra and in many cases they have a a PhD, meaning that this last step it's going to be mostly inaccessible to your average software engineering developer unless they want to go back to school, right?

So, if you are a software engineer, you will most likely switch between an AI user at the beginning and maybe an AI engineer, but in the end the AI engineer is a traditional engineer with a bit of an AI touch. We will dive a bit deeper into this AI engineer later in this video, but when you see AI fluency in job postings, this is what they mean. Okay, now, when it comes to technical interviews, certain companies won't even allow you any hands on coding.

So, you're going to be presented with a problem and they'll tell you, hey, you have to solve this problem and sometimes it's even a debugging problem, which is a bit counterintuitive, why would you prompt cloud code to debug something when most of the times it's easier for you to just dive in, right? But yes, I can give you dozens of examples from our menties in the last couple of weeks when they were faced with a problem in a technical interview and they were told the only thing they can do is to pump cloud code or a model of choice to solve the problem.

So I guess technical interviews feel like that these days. Now, both corporates and startups are demanding AI skills, which is tricky, right? Because you would believe that AI startups are the one obsessed and corporate are a bit slow, slower to move. And they both have different reasons, right? For corporates, yeah, they do want speed, but they also want stability too, because they have an existing business, right? So for them you are considered more of a human insurance against AI and also the developer that will probably be bring AI to the company.

So, if you interview with one of those, don't put all your focus sonio AI skills, because they're still interested in your code quality and testing and AI code reviews skills. So you want to talk about how much you use AI for your coding, but you also want to underline your traditional software skills. This is why companies are also asking for deep expertise in SV and software development best practices when you check job postings. Like strong understanding of web fundamentals, including performance, etc. Deep expertise in React Next JS and TypeScript and even strong computer science fundamentals. And I know you're probably thinking right now, dude, is this all the stuff I need to know just to get a software development job? And the answer is, yes, folks, this is these are the times we live in. Now, on the other hand, startups are obsessed with using AI to ship faster. Okay, they're all about token maxing and cloud code and velocity, right? So if you talk to these people about like SV best practices, you will get rejected on the spot. Here, you have to, you know, better keep most of your conversation about like context engineering and harness and handy AI coding tricks. The problem with these startups is that even if you get an offer to work there as a software engineer or an AI engineer, but the problem is they want to run fast. Many times they don't know where they are running towards. So it's like you're going to ship a lot of code, but throw away a lot of code, there's going to be a bit of chaos and instability. But if you are in for that, then I guess you can make it work. And by the way, there is an 100% chance that you will get this question in your next tangular interview, and that question is, how are you using AI in your daily workflow? And if your answer is, I know, I just asked cloud code everything I own in plan mode, you know, and then I delete to implement that plan and maybe I check that plan. Then you might want to elaborate a bit because we want to stand out, because that's what everyone is saying. I would prepare these questions before you go into an interview. For example, which AI coding tools do you use on a daily basis? Right, how do you integrate them into your workflow? How much of your code is written with AI, or what are the main pitfalls to watch for?

when coding with AI. And here you should mention common AI mistakes and code smells. And finally, how do you make sure that AI code quality stays consistent? That's the hardest thing. And also the most interesting question when we talk about AI coding. An example would be, well, by designing stable interfaces and having AI implement against those. Now, on to software engineering reality number two, the senior only market, it's ironically being flooded by fake seniors.

What happens is most job postings are looking for senior devs and mid and junior devs are basically cut out of the market, right? Here, we're looking at about 2,200 results, and we see that most of them are actually looking for senior engineers. But as soon as we filter out for entry level only, we get about 226 results. That's 10 times less junior and mid-level positions rather than senior level.

It's a senior's market and it's been like that for years. So what's happening? Well, developers are gaming the system. They're using AI to create fake senior resumes. And that's not unique to software engineering, everybody's doing it because it's so easy. Hey, GPT, just generate me a resume with like senior, senior, senior, full stack, whatever. Gardner predicts that by 2028, one in four candidate profiles worldwide will be fake.

I still don't know how HR is going to adapt to this, but we are going through interesting times. One in four people, it's faking stuff when it comes to job hunting. So what's happening is companies are reporting a drop of quality in candidates and job applications, specifically in software engineering. And many companies I've been talking to report similar situations to this one where they get hundreds of thousands of resumes that are either fake or spam.

And even when they get some candidates in and they interview them, many of them have no idea of any of the skills they added on the resume. My advice is that whatever you mention in your resume, make sure you are able to defend it in a technical interview. Now, you don't have to be proficient at it, but you do need to know about it and know how to explain it. And the best way to do that is by knowing what your gaps are first.

This is exactly what these free Tinker assessment that Bug and I built will do for you, you can check it out in the LinkedIn comments. Onto software engineering reality number three, technical interviews are not hard, they are insane, they will make you over break you. The thing is the margin of error in technical interviews, it's tiny these days. One vague answer and it's enough to get you rejected.

And the crazy part is that you will get questions all over the stack, doesn't matter what you interview for. You can go to an interview and get a question about micro front ends. Then you got to get the question about backend scalability and then in the next interview, you're going to get the questions about JavaScript fundamentals or browser details that you've never probably heard about. For example, corporates will have you go through eight stages interview processes and then they'll just reject you with a generic email.

Or startups will tell you, hey, you're amazing, you are the perfect fit, you know, I'll call you tomorrow, we want to hire you. And then they just ghost you or they tell you that they're run out of funding because they're just a serious just trying to raise money, etc. So, in order to stay sane in this market, you need a lot of patience and you cannot take things personally. Because the truth is, most companies don't know how to lead technical interviews.

Half of your interviews are going to be completely random, a front and interview, they start asking you backend scalability questions. Which is exactly why you should not blame yourself so high when you get rejected, just take notes and look at what you could improve in the next one. And here is a senior developer tip, if you're interviewing right now, don't get emotionally invested in any company or interview process.

Until they extend a written offer, you are still unemployed. That's exactly the reason why most developers crash mentally and emotionally when they are job hunting, because they get too emotionally involved. It's better if you simply have no expectations at all and you just keep job hunting until you get an offer. Here is some more OG technical interview advice, only apply to jobs posted in the past 24 hours.

With less than 100 applicants preferably, anything else, they won't even get to see your resume. Avoid taking on tasks at all cost. Don't accept any technical assignment without first talking to a real person. You don't commit to them with your time and effort if they don't commit to you. Also avoid automated or AI interviews, video interviews, etc. These people are just farming applications. They won't reach out to you in most cases.

Do them unless you have nothing else going on and you want to practice some interview questions with a robot. That's the only reason you should do them. And I guess if you're a job hunting, you don't have any spare time. Startups might have job expectations that you can't or don't want to fulfill. Like, hey, you know, you now that you can use AI, you can do like three jobs for the pay of one and work extra, work in the weekend and etc.

Maybe you are better off just avoiding these companies altogether. Know your resume by height and have some stories ready. I mean, you can use chat GPT to prepare them, but I would say write them yourself. Think about them. You know, this is a staff that you will talk about with a real human in every single interview you're going to have. So it's worth it to put like two, three, even four hours, okay, to prepare those things because you're going to use them over and over again.

And unless specifically asked by a human, do not tailor your resume. Build a resume for the market, not a specific position. Everybody's using these AI apps to change the resume, to fit the the job posting. And people will realize, hey, you know, this resume looks like like exactly the job posting, it doesn't really tell me anything about the person itself. A better strategy when everybody's a copycat is to be original.

Some more advice, structure your days. Okay, up skill and interview prep early in the morning when you're high energy. And in the afternoons, you can do interviews. And very early mornings, you can apply to jobs to make sure you are the first one as soon as the job gets posted. You send your resume. Also, most of the job postings you will still find on LinkedIn. I don't like LinkedIn at all, but the jobs are there and indeed.

If you run out of job postings, then you can check more local platforms that depend on where you are based. It's different for every country. Interviewing, it's a numbers game. Everybody says that, but they don't really apply it. So, the more interviews, the better you will get, and the better you get at doing interviews, the less interviews you need to get an offer. But you need numbers, you need to get good, but you also need numbers.

Don't get emotionally invested in an interview process. We talked about this already, you're still unemployed until you have a signed offer in your inbox. The job market right now, it's a buyer's market, most of you know that already. But companies hold the power right now and they know it, but it's getting better, just hang in there and keep doing your best. Also, have a daily routine, hit the gym, catch up with friends.

Job hunting, it's incredibly emotionally draining. You need some stability to counter that. And one last senior developer tip, it's easier than ever to stand out from other candidates who are using AI for everything. You just have to be more specific in your resume, interview answers, etc. The more real experience and real stories you can add and quantify in your resume, in your LinkedIn, whatever, the more your resume will stand out and the better it will convert.

Now, moving on to software engineer reality number four. While some developers are living high, others are driving Uber just to get by. While their AI stocks are only climbing higher, right? AI bros and investors are choosing whether they should get a McLaren or a Ferrari and they are signing over tops. Anyway, meanwhile, most software engineers are one email away from a financial disaster, getting laid off is the equivalent of switching social class.

Because software engineers who are employed are scared because the market is bad, so they hold on to their jobs. Meaning that if you get laid off, it's way harder to get a job because there simply aren't that many open positions. It's a low higher, low fire job market. So there's not a lot of movement going on. Which means, in some cases, some developers are even forced to get any kind of job they can just to pay bills, like this one who went to work for a cashier at the 7-Eleven and it's barely surviving.

Other people that I had calls with, you know, after a while, they tell you, hey, I've been driving Uber or doing food delivery because they ran out of savings. they assume hey he's going to take me six months to find a new job. I'm going to, you know, go on a sabbatical for 3 months and then they come back and he's like, wow, the market is totally different, etc. The market is totally different, etc.

One thing I'll tell these developers it's not a shame to do something else to make some cash and earn some money while you're looking for a developer job. But you cannot, and this is the difficulty in those situations, you cannot let the other job distract you. You have to keep your eyes on the ball, keep improving your interview skills and keep going at it. Other devs are trying to bypass completely the job market by becoming their own boss and building side hustles.

And of course, all this AI micro sass influences living in Bali, made them believe that code was the bottleneck and now that everybody can code, you can go out and with AI, put a solution out there in the market that's better than anyone. Now, I made another video about this on the channel. In reality, look, launching a successful micro sass company, it's way harder than getting a software development job and then these Googles want you to to know.

They don't usually make money with their their software products, they make money either, you know, selling courses to teach people to to create software products that they themselves cannot create. Or it's simply because they have a huge reach. What happens is when people try this and they vibe code micro sass fails, they have to go back to the market with an even bigger gap in their resume and less money in the bank.

On top of this, less students are choosing computer science careers. Here it says that enrollment felt by 8.1%. And my guess is that it will keep them falling. I mean, if you're constantly being told that AI will replace software engineers, then why even bother becoming one? I mean, students are rational human beings. students are rational human beings. Anyway, the sad part is that at the same time, the market is slowly getting better.

Software engine openings have been continuously growing throughout the year and we've seen this in our own mentorship program. Where students kept landing amazing offers. Despite the competitive job market. So I would say don't lose hope. soon be over and things will get back to normal, but you have to be ready. My advice to junior and mid-level devs is that sometimes you win just by surviving the storm.

So, as I said, just hang in there and wait for the dust to settle. And by the way, folks, if you like Bogdan and me to help you level up to a confident, competent, AI ready senior engineers with personalized mentorship, senior mental models and a clear road map, then check out the link in comments. On to software engineering reality number five. Pure front and rolls are being replaced by full stack ones.

I've talked about this trend a few months ago in this channel, but this is data from two.io. They track software engineering openings worldwide and specifically across the US and you can clearly see that there are at least 20 times more back end and full stack openings than pure front end roles. Which honestly makes no sense because it's the front end that's usually most labor intensive when building a software product.

So, what's happening? Because even for front end positions, they are also asking for full stack skills, a strong generalist or full stack capability or infrastructure knowledge and I just pulled out this job requirements from these front end engineer jobs such in San Francisco. Now, people speculate that this is because of AI, because you can use AI, you don't need a specialist, but in reality, I believe this is simply a cost cutting measure.

Companies are taking advantage of the market. So what they do is to hire a generalist and make them work extra hours instead of paying for two or three specialists and they tell them, hey, you know, just use AI and do the work that two three developers used to do in the past. And my prediction is that look, front end is core to marketing. Marketing is the game of standing out. Standing out is becoming way more way harder as AI it's inundating the web with its mediocre web content and stuff.

So companies will hire developers who can do better stuff than AI. And of course, that includes front end developers, so I really believe this trend will ui voice, and front end developers will be in huge demand in the near future. And another senior developer tip here it's, look, even if you plan to stay a front end developer, learn full stack skills. And the trick here, it's, look, you don't need the same level of depth we're mastery that you have in the front end, where you have to know all the edge cases, how the browser works, front end architecture.

You don't need all that in the back end, but you do need to know back end fundamentals and some mental models, and you need to be able to implement them at least once yourself. And if you want me to make another video more in depth about this one, then let me know in the comments. And again, if you want an easy way to assess your full stack skills, then remember the angle assessment, Bug Knight put together for you, check it out in the comments.

On to software engineering reality number six, AI engineer positions are popping up, but is this hype or reality? And when I say AI engineer, I'm not referring to a PhD level machine learning engineers, like the ones that work at AI companies like Anthropic or Open AI, developing AI models. Because these positions usually require a PhD in a relevant field to AI, which is either mathematics or statistics and knowledge that it's kind of out of reach for the great majority of software engineers out there.

What I'm referring to, it's AI engineer job postings like this one that are recently popping up on on LinkedIn, which require experience building and deploying LLM powered applications. I actually bumped into these job postings by analyzing the full stack job market and I was surprised to see so many of this AI and applied AI engineering positions. And then I realized that most of these positions are sponsored, meaning that companies are paying to advertise these jobs on LinkedIn.

And my guess is that most AI software engineer positions come from one, either series A or B startups with like six to 12 months of runway. That want to hire people fast. Or late stage companies. AI engineer and just ride the wave or just stay a traditional software engineer. And I'm saying, hey, no, you should not switch careers. And yes, you should expand your skills, but stay a software engineer first.

And here is why switching to AI engineer, it's risky right now. And I know there's a lot of influencers saying, hey, you should just jump in. Traditional software is dead, AI engineer is the future, they can't even define what AI engineer actually means. But yeah, we will just move on. The thing is, AI engineer can still be just a fad. When the AI bubble deflates, the today AI engineer job will be called, you know, back and engineer on AI features within 18 months.

Because in the end, if it's just calling an API that calls, I know, open AI or entropia or whatever API, then that's not an AI engineer. The other thing is, if AI investment disappears tomorrow, then pure AI engineers at startups will be laid off before general SBEs. Because companies will keep developers who are in critical areas, database, maintenance and APIs and infrastructure, not just people that are making API calls through to LLMs.

And finally, the AI engineer will saturate quickly. Why? Because most developers can learn reg and highness engineering in a weekend. Honestly, there's there's no magic to it. And so the differentiator will be deep software engine knowledge and judgment, you know, knowing when not to use AI. How to evaluate it properly and how to bring it to production. That comes from deep tangled mastery and traditional soft engine experience.

So no, you should not just switch careers and pivot to AI engineers, just like some people a few years ago, they switch to like crypto development. And then crypto went down and then they had to go back to full sack engineering. But you should add some of those AI skills to your skill set. With that being said, folks, thank you so much and I will see you in the next one.