Comment by xtracto | Hacker News Reader

Comment by xtracto | Hacker News Reader

Parent Full thread

You made me login to reply :) , thanks.

Back in the early 2000s when I was finishing my Software Engineer BSc degree I saw the choice of becoming a "generalist" vs becoming a "specialist". I actually liked EVERYTHING technology wise: From Neural Networks to Game development (graphics with OpenGL) to algorithms, Web development, to Java JNI, assembler and whatnot. I couldn't see myself focusing in one thing.

Fast forward to 2025, I'm 44 years old and have been 24 years in the industry. In the last 5 years I've had 3 jobs: One, helping a startup move form a non-scalable monolith system (ruby) to a very scalable microservices one. I was CTO of a crypto-exchange company, building ECS/nodejs based microservices and then an App (React Native). And right now I am helping some young guys in a startup doing AI based Tax reconciliation (helping exporting companies recover their VAT).

In my opinion, right now is the BEST moment to be a developer. Coding with Cursor is magic. Implementing an API in python with FastAPI is so freaking easy and quick. I don't have to worry about recalling a lot of details, but mainly think on problem solving.

I have the hypothesis that the people that are struggling are the "specialists". Suddenly with AI it doesn't matter that you know the in and outs of Java, Hybernate and the whole stack. There's more value in solving problems. I am happy that I chose the "generalist" path. I think AI will reduce the demand the "specialist" skillset.

I am very much a generalist and this comment checks a lot of boxes for me. If you find yourself pigeonholed into some niche (which can be super profitable), and that niche disappears, you're not really left with much to work with other than a complete career change.

this is a factor for me. I've always had to pick up new languages frameworks and skills on-the-job. but today hirers are seeking only really niched down specific experts, and i suspect i am filtered out automatically by not matching that, with no regard to the fact that it's not that big of a deal for an experienced engineer to pick up something new

you are right in that as a generalist i have no fear to start a job in any new tech stack, no matter how unfamiliar. but with hundreds of applicants to every job there is always someone more experienced with the tech stack used there. to win as a generalist you kind of have look for jobs in niches where only few people apply. i am struggling as a generalist.

I found much better success on wellfound.com (formerly angel.co) than LinkedIn. Much more startup and generalist-focused, whereas LinkedIn only produced through recruiter recommendations.

As a generalist "master of none" who also graduated from the same era, this is really reassuring. I use a lot of technologies but not enough to consider myself an expert in any of them. At the end of the day I can pipeline them into a useful tool or product.

I also recognize when AI is getting the answers wrong. LLMs are great at giving you general, well documented answers. For the moment it doesn't have the foresight to tackle complex systems. And that is where a specialist can really shine. But the world doesn't need a lot of answers to complex problems when most of the time a general one will do.

Reply on news.ycombinator.com