Tech job postings dropped 34% between January 2023 and March 2024, according to Lightcast’s 2024 Labor Market Analytics Report, even as the broader U.S. unemployment rate held below 4%. The boom did not slow down. It inverted.
If you are a mid-career engineer right now, that number should stop you cold.
So which is it? A boom or a bloodbath? Depending on your skill stack and your job search strategy, the answer is one or the other. There is almost no in-between anymore. And most people are on the wrong side of that line without knowing it.
What the Headlines Are Getting Wrong
The mainstream narrative is that tech hiring is “cooling.” That is the polite version. The accurate version is that a specific segment of the tech labor market collapsed, and it collapsed fast, while a different segment accelerated with almost no fanfare.
Generalist software engineering roles, the kind that were minted by the thousands during the 2020 to 2022 hiring surge, are the ones evaporating. According to Layoffs.fyi, over 262,000 tech workers were laid off in 2023 alone. Meta, Amazon, Google, and Microsoft shed tens of thousands of positions across a single calendar year.
But here is the number that matters: companies added AI/ML engineering roles at a rate of 74% year over year in the same period, according to Indeed Hiring Lab’s 2024 data. Cloud security specialists and data engineers saw comparable demand spikes. The job market did not disappear. It bifurcated, brutally and without announcement.
This is not a cyclical adjustment. It is a structural reset. Full stop.
The Mistake That Is Killing Your Chances
How many applications have you sent this quarter, and how many of those led to a first call?
If the ratio is worse than 1 in 20, the problem is almost certainly not your resume formatting. It is positioning. Specifically, you may be marketing yourself as a generalist in a market that has stopped paying generalist wages.
Here is what that looks like in practice. Call him Marcus. He is 35, eight years of experience at a Series B SaaS company in Austin, Texas, $142,000 base salary as a senior backend engineer. Competent across Python, Node, and some React. Solid fundamentals. On paper, a strong candidate.
Marcus was laid off in February 2024 when his company reduced headcount by 30%. He started applying. After eleven weeks and roughly 80 applications, he received three offers. The highest came in at $104,000. The roles were not bad roles. But the market had repriced his profile by $38,000 because nothing in his portfolio screamed specialization. He had broad exposure to AWS but no certifications, no public projects, and no narrative around cloud infrastructure. Recruiters read him as a generalist. That is the box they filter out first.
His story is not rare. It is the median experience right now for mid-career engineers who built careers during an era when full-stack flexibility was rewarded.
Warning: If your most prominent skills section leads with languages and frameworks rather than domain outcomes, such as “reduced ML inference latency by 34%” or “migrated three microservices to AWS ECS with zero downtime,” you are being read as a commodity. Commodity gets filtered. Specialist gets called.
Why This Is Happening (And Why Smart People Missed It)
The 2020 to 2022 surge was real. Companies were scaling digital infrastructure at a pace that had no historical precedent. Engineers were hired broadly, compensated generously, and asked to move fast across whatever was needed. Flexibility was the asset.
Then two things happened simultaneously. Interest rates rose sharply, which forced every VC-backed company to start justifying headcount in ways they had not for three years. And large language models became production-ready, which allowed companies to handle generalist software tasks with smaller teams and AI-assisted tooling.
Are you competing in a generalist market without realizing it? That is not a rhetorical question. The engineers I hear from most often are genuinely surprised that their profile is not landing. They are not bad engineers. They are positioned for a market that existed two years ago.
The companies that are still hiring aggressively, and there are many of them, are hiring for outcomes in specific domains. They want someone who has shipped an ML pipeline, hardened a cloud environment, or built data infrastructure at scale. They are not interested in potential. The runway for potential closed when interest rates went up.
Who Is Actually Getting Hired
Let me be direct about this. The engineers clearing interviews in 2024 share a few characteristics that have nothing to do with raw IQ or years of experience.
They have a visible specialization. Their LinkedIn profile, their GitHub, and their resume all tell the same story about one domain. They do not read like a Swiss Army knife.
Does your LinkedIn profile right now reflect a specialization, or does it read like a collection of every tool you have ever touched? That is the question a recruiter is answering in about 12 seconds.
They have external proof. Not just listed skills. Actual artifacts: a deployed project, a certification from AWS or Google Cloud, a published write-up, a GitHub repo with real commits. Something a recruiter can click before they pick up the phone.
Their salaries have not compressed significantly because supply in their lane is genuinely thin. MLOps engineers with two or more years of production experience are still clearing $160,000 to $190,000 at mid-sized companies, according to Levels.fyi data from Q1 2024. Cloud security engineers are in similar territory. These are not unicorn roles. They are specialist roles in lanes that generalists have not flooded.
This connects to a broader shift I have written about before: enterprises moving away from bloated, generalist-built proprietary stacks toward leaner, specialized infrastructure. If you want to understand the employer side of this equation, why enterprises are ditching proprietary software in 2026 is worth fifteen minutes of your time.
Did You Know: According to Glassdoor’s 2024 Tech Compensation Report, MLOps and cloud security roles saw median salary increases of 11% and 9% respectively between 2022 and 2024, even as median software engineering salaries fell 6% across the same period.
Quick Check: Pull up your resume right now. If your most prominent section leads with languages and frameworks rather than domain outcomes, recruiters are reading you as a generalist. That is the first box they filter out. Specialization is not a buzzword. It is a price tag.
And if you are a newer engineer navigating this, the dynamic is even sharper. The bootcamp pipeline is still producing generalists into a market that has closed that door. Too many bootcamp grads, one exit lays out exactly what that correction looks like and what to do about it.
Your Next 3 Steps
Most people get this wrong by treating job searching as a volume problem. It is a positioning problem. Here is how to fix it before Friday.
Step 1: Audit your last three projects against real job postings this week. Open LinkedIn Jobs or Levels.fyi and pull five current job postings in MLOps, cloud security, or data engineering. Read the requirements. Now look at your last three work projects and identify every touchpoint, however minor, with AI, cloud infrastructure, or data pipelines. Rewrite three resume bullet points around those touchpoints using outcome language: latency reduced, cost cut, pipeline scaled. Minor exposure reframed correctly reads as relevant experience. Do this before you send another application.
Step 2: Activate your network with a specific script, not a vague check-in. Identify five hiring managers at target companies on LinkedIn. Do not send a generic connection request. Send a three-sentence message: one sentence naming something specific they built, wrote, or posted; one sentence stating your specialization and a result you delivered; one sentence asking if they have fifteen minutes in the next two weeks. That cadence, five messages, specific and brief, will generate more first calls than eighty cold applications. Do it this week.
Step 3: Commit to one specialization lane and create one public artifact. Pick MLOps, cloud security, or data engineering. Spend two hours this week either completing one milestone in the AWS Cloud Practitioner or Google Professional Data Engineer certification path, or deploying one small project to GitHub that demonstrates domain work. Link to it in your outreach. One clickable artifact changes how a recruiter reads your profile. It signals that you are already working in the lane, not just claiming interest in it.
The engineers thriving in this market are not smarter than Marcus. They are positioned differently. That is a fixable problem. The market cleared the generalists. The specialists are still getting calls.
That is not pessimism. That is the market telling you exactly what it wants.
