[danans]

> Someone asks you to add a feature to an existing program

While I empathize with the tone, even before AI the creativity was largely at the feature definition step, not in the implementation.

Outside of the very few computer scientists working on novel algorithms, the vast majority of software development is a mapping problem between the feature request and the mundane technical details, something repeatedly (and correctly) mentioned here in the context of FAANG algorithm fixated interviews. This has now largely been automated by LLMs

What is left is just creativity part - defining the use cases and features to develop in the first place. But the corollary is that software engineers that start after the requirements have already been defined are obsolete, which is a sobering thought for any of us in that vocation.

[tgarrett]

> Outside of the very few computer scientists working on novel algorithms,

It's a quite a bit broader than that: for instance most of science and engineering is heavily supported by simulations (very useful when the system you're considering doesn't have perfect spherical or cylindrical symmetry), and there is still tons of algorithm development going on. The world is vast, and thus so is the domain of programming.

And halfway through 2026, AI has become a very interesting and helpful partner in algo research too. If it does continue to pull away and zip off to ASI land, hopefully we can leverage the resulting magical technology and catch back up with it...

[danans]

> It's a quite a bit broader than that: for instance most of science and engineering is heavily supported by simulations (very useful when the system you're considering doesn't have perfect spherical or cylindrical symmetry),

That isn't the vast majority of traditional software engineering work, and arguably is better called applied physics or applied science. Super interesting though - and definitely uses programming as a core skill/tool - but it leans heavily into traditional engineering and science.

[tgarrett]

> That isn't the vast majority of traditional software engineering work, and arguably is better called applied physics or applied science.

Fair enough, and yeah definitions are always going to be somewhat fuzzy. Still it seems safe to assume there are also a lot of novel things going on in games, embedded, finance, AI itself of course... Generally I can't help but feel that we have only dipped our toes into the vast ocean of program space, and I'm curious what else is out there.

[ezoe]

Such a creative era was until early 1990s. Back then, we have to implement almost everything that is provided by standard library and popular frameworks we take it for granted for now.

[Chipshuffle]

Also typically new features resemble ones I implemented in the past and I won’t learn that much from implementing them. Often times I immediately know what to do and although I do enjoy typing I do prefer just having it done in way less time.

In the meantime I can concentrate on more interesting problems.

[pianopatrick]

Seems to me that feature requests are not all that unique either.

Seems to me you can get feature requests from user feedback.

I also expect someone will figure out a way to have AI "use" your software and suggest features

[slopinthebag]

This is not true. First of all, not all software is written in the context of a FAANG company with “feature requests”. Secondly, writing software is similar to the process of design, this comment reads like “the vast majority of handbag design is mapping problem between features and leather”, ignoring that both the design and implementation can be rewarding to work on. Eg. I’m working on a program for myself and the overall architecture of the program as well as some parts of its implementation are clever and compose well to make the codebase a joy to work in. I am not simply “mapping features to mundane technical details”. It is as much art as the skillfully hand-crafted handbag.

[danans]

> I’m working on a program for myself and the overall architecture of the program as well as some parts of its implementation are clever and compose well to make the codebase a joy to work in. I am not simply “mapping features to mundane technical details”.

You said it: you are working on a "program for yourself". Hobbyist craft programming will always be here. The question is what kind of software engineering will be paid for, and a career can be built on.

I don't see much of a market for pure software engineers anymore. You also need to be a product manager, scientist, or have some other domain knowledge adjacent to software that relates to the real world.

I say this with empathy for those who just enjoyed the craft of designing and building software, and thought that alone would provide them a livelihood and career in perpetuity, but have found a big chunk of what they loved doing (and getting paid for) overtaken by AI coding agents.

[slopinthebag]

What I said equally applies to commercial software as well. It's pretty much the only way to build software which is extensible, maintainable, mostly bug-free, and performant. That companies often churn out slop isn't proof that it's unnecessary, it's evidence that it is.

[horns4lyfe]

I always think that stuff is funny because it clearly comes from a place of having only worked at faang-level-esque companies. I’ve only ever worked at messy mid-size software companies and have never once had a legit product manager guiding feature work, it’s generally been up to the developers to figure out what needs doing

[danans]

> I’ve only ever worked at messy mid-size software companies and have never once had a legit product manager guiding feature work, it’s generally been up to the developers to figure out what needs doing

That's because the systems and problem spaces at FAANG companies are so large that you need(ed) a lot more division of labor to make anything work. The division of the technical side of the house between those focused on product management and those focused on building software allowed both to be done more effectively, but both roles have had a lot of their work overtaken by LLMs.

Also, From what I've heard, FAANG companies are going through big (and painful) transitions to align their staffing with the reality of what current AI tools are capable of.