Should I (or my kids) go into software engineering
Taking a stab at a question I get asked a lot.
April 25, 2026
So this is a question that I get asked quite a bit. In a world of increasingly competent AI should I / my kids / anyone pursue a career in software development in 2026. I think this tricky to answer.
2025 saw unprecedented layoffs in all flavors of tech. Here in 2026 things aren't much better. May alone looks like it is going to be absolutely terrible for people, and the job market they are facing is exceptionally tough. Salary expectations are trending down, and supply is out-pacing demand for talented software engineers.
People are arguing that a lot of companies aren't truly seeing AI productivity gains, and are just shedding COVID-era over hiring at the tail end of the free-money interest rates. Certainly some of this is true, but don't write this off as a blip. Technology leaders are already seeing the capabilities of AI today, and realizing that the way you staff and structure engineering teams in the new reality is going to be very very different than what it was before. Tighter, more senior teams, are seeing the most productivity gains. Even if token costs go up 2-3x (hell even 10x) they are still economical for a huge swath of software engineering tasks. This only gets better as models continue to improve, and the coding harnesses further improve. Look how far we've come in the last year alone.
So to return to the question at the top. There is going to be an over supply of talented and experienced software engineers for the next several years. It is also very unclear if companies will ever return to the massive team scales we saw through the last decade and a half. If the pattern of teams tightening around fewer more senior individuals, there will be fewer opportunities for new grads and young devs. There is a massive unanswered societal question of how do we get senior engineers without junior engineers, but it is unlikely the majority of companies will want to foot the bill to be altruistic. I am not advocating for this, but I do believe it will be the reality.
The narrative starting in the late 90s through essentially 2024 was that software engineering is a great career. 20+ years ago, even amongst my friends going into computer science along side me, many didn't really have a passion for it, but they / their parents thought it would be a good path to high paying jobs. That advice has been echoed consistently since then, often cited to folks in manufacturing jobs as a path forward when finding themselves automated out of a job - just learn to code. That is now looking like bad advice.
If you or someone you know is passionate about building software, turning their ideas into real products, they should absolutely still pursue computer science and software engineering. I still love what I do, frankly more than ever. Many of my peers and friends feel the same. There are always going to be opportunities for people with the drive to build things, whether its joining an organization or starting your own. AI lowers the barrier for the latter substantially.
I encourage people starting out to focus their self studies differently than they used to. While coding bootcamps and online training courses often focused on quick wins like standing up an app, building a functional UI, etc - all of these things are now trivially produced with frontier AI models. Instead focus on learning the fundamentals. How code executes through the operating system on the processor. Learn architectural patterns, how to build things that scale. Go deep on security, ensure you understand common pitfalls and how you design things defensively. Learn how to review software and ensure you can truly understand code not written by you (in this case AI).
So the question probably needs to get reflected back on someone wondering if software engineering is for them. Is this really something you want to do? If you've got a passion for building its now easier than ever. Go out and create products for yourself, your friends and family. If this brings you energy there is still a path forward in this industry. Otherwise probably not.