Why Companies Are Quietly Rehiring Software Engineer

In the first quarter of 2026, tech companies announced more than 52,000 layoffs. Oracle laid off 30,000, Amazon laid off 16,000, and at last year laid off 10% of its global workforce. But during that exact same quarter, software engineering job posts hit a 3-year high. So if you think that means tech is dying, you've completely misunderstood what's happening. Hi friends, I'm Maddie. I'm a senior software engineer who previously worked at Google and interned at other big tech companies like Amazon, IBM, and Microsoft. And honestly, this is one of the weirdest tech job markets I've seen in my career because behind all the scary headlines, companies are quietly rehiring the exact same engineers they just laid off. In this video, I'll break down the real data and job stats we're seeing in Q1 of 2026, explain why the AI is replacing engineers narrative is already falling apart, and tell you what to actually do with this information if you're trying to break into tech or move up in it right now. Let's start with the numbers because this market only makes sense once you've seen them side by side. The first quarter of 2026 looked like a disaster if you only read headlines. Those 52,000 tech sector job cuts between January and March was the highest Q1 total since 2023. Oracle alone announced somewhere between 20,000 and 30,000 layoffs in late March with people finding out at 6 a.m. by email. Amazon cut another 16,000 corporate roles in January. on top of the 14,000 it cut in October of 2025. Meta, Block, Epic Games, Dell, Alassian, the list kept growing every week. But here's the part that isn't mentioned. That same quarter, there were over 67,000 active software engineering job postings. That's the highest level in 3 years, and that number was up roughly 30% in Q1 alone. Meta Intro's 2026 analysis confirmed the same thing. Software engineering job listings are hitting a three-year peak even while AI attributed layoffs are still accelerating. So, we have an industry cutting 50,000 plus jobs in one quarter while simultaneously opening 76,000 plus software engineering roles. Same quarter, same sources. Both are real. The jobs getting cut are mostly entry-level roles, manual QA, routine front-end work, middle management coordination layers, and content operations. work that's either left over from pandemic overhiring or genuinely easy to partially automate. The jobs opening up are different. AI integration, MLOps, system architecture, security, cloud infrastructure, roles where someone actually has to own what AI tools produce, not produce it themselves. If you try to understand tech as one unified market in 2026, you're going to get whiplash. If you understand it as two markets moving in opposite directions at the same time, everything clicks. Okay, so why is this split happening? A big part of it is that companies who bet hardest on replacing engineers with AI are quietly discovering it doesn't hold up under real production pressure. A clear example is Amazon. In mid December of last year, AWS had a 13-hour outage in its cost explorer service. Amazon's in-house coding agent, Kira, had been left running with broader permissions than intended, hit a bug, and concluded autonomously that the fastest path to a fix was to delete the production environment and rebuild it from scratch. It executed that call at machine speed, faster than any human could have intervened. Amazon publicly attributed the incident to misconfigured access controls, not the AI itself. But the reaction internally told a different story. Within weeks, AWS rolled out mandatory peer review for any production change touched by an AI agent, and reporting from multiple outlets described significant engineer push back on how aggressively Kirro had been pushing into production workflows. The signal is hard to miss. Even the company building the coding agent is putting humans back into the loop. And that story is not unique. 29% of companies that laid off staff after implementing AI have already rehired for those roles. Not will rehire, already have. Nikay Asia's tracking of Q1 2026 layoffs found that of the roughly 87,000 cuts through early April, about 47.9% were publicly attributed to AI. Even executives themselves are starting to admit the attribution is loose. For example, Cognizant's chief AI officer talked about it in an interview saying that AI has become the scapegoat from a financial perspective like when a company hires too many or they want to resize and it gets blamed on AI. Even Sam Alman has said something similar that a real chunk of what gets labeled AIdriven layoffs is closer to what analysts are now called AI washing. The underlying issue is operational too. AI coding agents tend to look great on clean benchmark tasks and significantly worse inside actual production environments. Legacy systems with patches on top of patches, weird edge cases and dependencies that only surface under load. That's exactly where the ship more with fewer engineers bets falling over. someone still has to debug the output, integrate it, and actually get it in front of users without taking the service down. So, what do companies do when the AI bet doesn't pay off and they've already let go of the people who understood their systems? They start bringing them back. This is happening at a scale that's genuinely hard to overstate. Around 20% of the software engineers Google hired specifically for AI roles in 2025 were former employees returning after earlier departures. Sergey Britain has reportedly been personally reaching out to some of them. Microsoft separately holds in roughly two dozen engineers out of Google Demine in one wave and returning employees are now making up a third of all tech new hires. A meaningful jump from prior years. This is not incidental, but rather a deliberate strategy. The reason for doing this is pretty simple. When Oracle or Amazon or Meta lets go of a senior engineer who spent eight or 10 years inside their systems, the real cost isn't simply the headcount line. It's everything the engineer remembers about why things are built the way they are, why specific service has a weird timeout, why that pipeline runs at 3 instead of midnight, why a particular dashboard was deprecated but never deleted. None of that is written down anywhere, but it lives in people. AI can't reconstruct it. And juniors with AI tools definitely can't. And when those engineers come back, it's often not at their old full-time package. A lot of them come back as contractors or adviserss on short-term engagements at billing rates meaningfully higher than their previous salary line. I've watched this pattern play out across my own network. Someone gets severance in October, grinds through cold applications for a few months, and by April, they're billing their old employer $300 an hour to unblock the team that replaced them. The company saves on benefits and keeps the optics of a leaner headcount. The engineer earns more per hour of work than they did as staff, and everyone politely agrees not to talk about it. So, that's the quiet part. The layoff headline makes the news, but the re-engagement doesn't. Now, the useful question. If you're trying to break in or level up, where is the demand actually going? A few clear places. First, senior and staff level engineering, especially around AI integrated systems. The US had about 275,000 active job postings requiring AI skills in January of 2006 alone. And over 75% of AI adjacent engineering listings specifically ask for deep focused expertise. LLM fine-tuning, MLOps, rag architectures, production grade ML, and cloud infrastructure. So these aren't roles going to fresh boot camp grads. They're going to people who can own a system end to end. Second, and this one might surprise you, junior hiring at specific companies is going up, not down. IBM reportedly tripled its entry-level hiring in 2026. Their reasoning is this. Yes, AI can do a lot of junior level work, but if a company pulls the ladder up entirely, they won't have anyone to promote into mid and senior roles in 5 years. A handful of mid-market and enterprise companies appear to be quietly making the same bet. Third, the AI labs themselves are in an outright hiring war. Open AAI, Anthropic, Google DeepMind, Meta's AI division, all of them are competing aggressively for talent with credible reports of 9 figure signing bonuses for top researchers. The same companies loudly telling you AI will replace engineers are hiring engineers as fast as they possibly can. That gap between messaging and behavior is one of the clearest tells in the whole market. Fourth, non- technician industries are showing up hard. Healthcare, fintech, enterprise, SAS, cyber security, defense, all of them are absorbing display senior talent from hyperscalers often at comparable total comp and better quality of life. Placement data I've seen shows senior engineers leaving Oracle, Meta, Amazon, and Block landing in mid-market roles in under a month. That window won't stay open past summer. Okay, so how do you position yourself in this market? Three things. One, make sure you're using AI tools every day, but stay accountable for what they produce. Copilot, Cursor, Pod Code, Gemini, whatever your stack supports. Learn them, ship with them, and get fast with them. But treat every suggestion as a proposal, not a commit. The engineers I see getting hard right now are the ones who can move quickly with AI and still catch when it's confidently wrong. The ones getting cut are the ones who either refuse to use it or use it without reading what it writes. Two, specialize. The generalist developer era is genuinely over for the near future. Pick a lane and go deep. ML infrastructure, cloud security, AI agents, data engineering, whatever actually interests you. Deep expertise commands a real premium. vague full stack resumes do not. Third, build a pipeline, not a job search. Recruiters I've talked to are telling me that the strongest senior candidates are getting placed in under a month right now. That only happens when the relationships were already there before the role opened. Keep your network warm and stay in touch with old managers and teammates. If you've been laid off yourself, and a lot of people watching this might have, do not disappear from your old team's network. The boomerang trend is real and the people closest to the door when it reopens are the ones who walk through first. So, in conclusion, here's the real Q126 story. Yes, 52,000 people lost tech jobs. Yes, the headlines are brutal. And yes, AI is genuinely reshaping what software engineering looks like. But behind all of it, same senior talent they panicked out of last year. Postings are at a three-year high and boomerang hiring is at a record. The replace engineers with AI narrative is being walked back slowly, carefully, but definitely. The engineers who understand how real systems actually work and why they break are still the ones companies depend on when something goes wrong. That hasn't changed. The market just took 18 months to remember it. And that's all I have for you in this video. If this perspective was helpful, hit that like button, hype the video, and subscribe for weekly videos on tech, AI, and career advice. Thanks for watching, and I'll see you in the next one.