Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: SWEs how do you future-proof your career in light of LLMs?

news.ycombinator.com

Post reply on HN
Print view View original

961–970 of 1001 posts

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#961

Post by indrora »

taylodl wrote: post #832
indrora wrote: post #316

Earlier quoted context omitted.

> Bottom line, you don't see too many assembly language developers anymore. And where you do, no LLM is going to replace them because they are working in the dark mines where no compiler has seen and the optimizations they are doing involve arcane lore about the mysteries of some Intel engineer's mind while one or both of them are on a drug fueled alcoholic deep dive.

Out of curiosity, who does assembly language programming these days? Back in the 90s the compilers had learned all our best tricks. Now with multiple instruction pipelines and out-of-order instruction processing and registers to be managed, can humans still write better optimized assembly than a compiler? Is the juice worth that squeeze? I can see people still learning assembly in a pedagogical setting, but not in a…

Assembly is still relatively popular in the spaces of very low level Operating System work, exploit engineering, and embedded systems where you need cycle-accurate timing.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#962

Post by simianparrot »

paulcole wrote: post #893
simianparrot wrote: post #264

Earlier quoted context omitted.

I'm future proofing my job by ensuring I remain someone whose brain is tuned to solving complex problems, and to do that most effectively I find ways to keep being engaged in both the fundamentals of programming (as already mentioned) and the higher-level aspects: Teaching others (which in turn teaches me new things) and being in leadership roles where I can make real architectural choices in terms of what hardware t…

Nobody who's been replaced by a parrot thought they were going to get replaced by a parrot. Where's your espoused intellectual curiosity and mental ambidextrousness when it comes to LLMs? It seems like your mind is pretty firmly made up that they're of no worry to you.

In another comment I briefly mention one use case I have been implementing with LLM's: https://news.ycombinator.com/item?id=42434886

I'm being vague with the details because this is one of the features in our product that's a big advantage over competitors.

To get to that point I experimented with and prototyped a lot using LLM's; I did a deep dive into how they work from the ground up. I understand the tech and its strengths. But likewise the weaknesses -- or perhaps more apt, the use cases which are more "magic show" than actually useful.

I never dismissed LLM's as useless, but I am as confident as I can possibly be that LLM's on their own cannot and will not put programmers out of jobs. They're a great tool, but they're not what many people seem to be fooled into believing that they are; they are not "intelligent" nor "agents".

Maybe it'll be a bit difficult for juniors to get entry level jobs for a short while due to a misunderstanding of the tech and all the hype, but that'll equalize pretty quickly once reality sets in.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#963

Post by purple-leafy »

pdimitar wrote: post #941
purple-leafy wrote: post #936

Earlier quoted context omitted.

There are some very narrow minded, snarky people here, “doesn’t pass my sniff test” okay Einstein, you do you

Surely, you actually mean those who extrapolate from a few niche tasks to "OMG the programmers are toast"? Fixed it for you. ;)

Sorry I don’t mean to be snarky :) I think there is a happy middle ground between

“AI can’t do anything, it sucks!”

and

“AI is AGI and can do everything and your career is done for”

I teeter along the spectrum, and use with caution while learning new topics without expertise.

But I’ve been very surprised by LLMs in some areas (UI design - something I struggle with - I’ve had awesome results!)

My most impressive use case for an LLM so far (Claude 3.5 Sonnet) has been to iterate on a pseudo-3D ASCII renderer in the terminal using C and ncurses, where with the help of an LLM I was able to prototype this, and render an ascii “forest” of 32 highly detailed ascii trees (based off a single ascii tree template), with lighting and 3 axis tree placement, where you can move the “camera” towards and away from the forest.

As you move closer trees scale up, and move towards you, and overlapping trees don’t “blend” into one ascii object - we came up with a clever boundary highlighting solution.

Probably my favourite thing I’ve ever coded, will post to HN soon

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#964

Post by pdimitar »

purple-leafy wrote: post #963
pdimitar wrote: post #941

Earlier quoted context omitted.

Surely, you actually mean those who extrapolate from a few niche tasks to "OMG the programmers are toast"? Fixed it for you. ;)

Sorry I don’t mean to be snarky :) I think there is a happy middle ground between “AI can’t do anything, it sucks!” and “AI is AGI and can do everything and your career is done for” I teeter along the spectrum, and use with caution while learning new topics without expertise. But I’ve been very surprised by LLMs in some areas (UI design - something I struggle with - I’ve had awesome results!) My most impressive use c…

I absolutely agree that it's a spectrum. I don't deny the usefulness of some LLMs (and I have used Claude 3.5 lately with great success -- it helped me iterate on some very annoying code with very niche library that I would have needed days to properly decipher myself). I got annoyed by the grandiose claims though so likely got a bit worked up.

And indeed:

> “AI is AGI and can do everything and your career is done for”

...this is the thing I want to stop seeing people even imply it's happening. An LLM helped you? Cool, it helped me as well. Stop claiming that programming is done for. It's VERY far from that.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#965

Post by schappim »

I think what we're seeing echoes a pattern we have lived through many times before, just with new tooling. Every major leap in developer productivity - from assembly to higher-level languages, from hand-rolled infrastructure to cloud platforms, from code libraries to massive open-source ecosystems - has sparked fears that fewer developers would be needed. In practice, these advancements have not reduced the total number of developers; they have just raised the bar on what we can accomplish.

LLMs and code generation tools are no exception. They will handle some boilerplate and trivial tasks, just like autocompletion, frameworks, and package managers already do. This will make junior-level coding skills less of a differentiator over time. But it is also going to free experienced engineers to spend more time on the complex, high-level challenges that no model can solve right now - negotiating unclear requirements, architecting systems under conflicting constraints, reasoning about trade-offs, ensuring reliability and security, and mentoring teams.

It is less about "Will these tools replace me?" and more about "How do I incorporate these tools into my workflow to build better software faster?" That is the question worth focusing on. History suggests that the demand for making complex software is bottomless, and the limiting factor is almost never just "typing code." LLMs are another abstraction layer. The people who figure out how to use these abstractions effectively, augmenting their human judgment and creativity rather than fighting it, will end up leading the pack.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#966

Post by amrocha »

aprilthird2021 wrote: post #943
amrocha wrote: post #626

Earlier quoted context omitted.

I don’t think I’ve ever heard of a developer who enjoys refactoring a messy, bug ridden code base. This is the reason why rewrites happen, nobody wants to touch the old code with a ten foot pole and it’s easier to just rewrite it all. So if you turn the entire job into that? I don’t think skilled people will be lining up to do it. Maybe consulting firms would take that on I guess.

You can't just rewrite all of some of the most profitable software systems out there. Banks, HFTs, FAANG, these cannot just be rewritten

Sure, that’s why they pay a lot and hire quality engineers to maintain their codebases. And wouldn’t be ok with AI slop.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#967

Post by gjadi »

taylodl wrote: post #832
indrora wrote: post #316

Earlier quoted context omitted.

> Bottom line, you don't see too many assembly language developers anymore. And where you do, no LLM is going to replace them because they are working in the dark mines where no compiler has seen and the optimizations they are doing involve arcane lore about the mysteries of some Intel engineer's mind while one or both of them are on a drug fueled alcoholic deep dive.

Out of curiosity, who does assembly language programming these days? Back in the 90s the compilers had learned all our best tricks. Now with multiple instruction pipelines and out-of-order instruction processing and registers to be managed, can humans still write better optimized assembly than a compiler? Is the juice worth that squeeze? I can see people still learning assembly in a pedagogical setting, but not in a…

ffmepg is famous for using asm. For exemple : https://news.ycombinator.com/item?id=42041301

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#968

Post by mariconrobot »

This is an interesting challenge, and I think it speaks to a broader trend we’re seeing in tech: the tension between innovation and operational practicality. While there’s a lot of enthusiasm for AI-driven solutions or blockchain-enabled platforms in this space, the real bottleneck often comes down to legacy infrastructure and scalability constraints.

Take, for example, the point about integrating AI models with legacy data systems. It’s one thing to build an LLM or a recommendation engine, but when you try to deploy that in an environment where the primary data source is a 20-year-old relational database with inconsistent schema updates, things get messy quickly. Teams end up spending a disproportionate amount of time wrangling data rather than delivering value.

Another issue that’s not often discussed is user onboarding and adoption friction. Developers can get carried away by the technical possibilities but fail to consider how the end-users will interact with the product. For instance, in highly regulated industries like healthcare or finance, even small changes in UI/UX or workflow can lead to significant pushback because users are trained in very specific processes that aren’t easy to change overnight.

One potential solution that I’ve seen work well is adopting iterative deployment strategies—not just for the software itself but for user workflows. Instead of deploying a revolutionary product all at once, start with micro-improvements in areas where pain points are clear and measurable. Over time, these improvements accumulate into significant value while minimizing disruption.

Finally, I think there’s a cultural aspect that shouldn’t be overlooked. Many organizations claim to value innovation, but the reality is that risk aversion dominates decision-making. This disconnect often leads to great ideas being sidelined. A possible approach to mitigate this is establishing “innovation sandboxes” within companies—essentially isolated environments where new ideas can be tested without impacting core operations.

Ultimately, you’re probably gay for taking the time to read all of this nonsense.

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#969

Post by ta12653421 »

pdimitar wrote: post #821
dreamfactored wrote: post #714

Earlier quoted context omitted.

> LLMs don't write code like that As someone who has been doing this since the mid 80's in all kinds of enterprise environments, I am finding that the latest generation are getting rather good at code like that, on par with mid-senior level in that way. They are also very capable of discussing architecture approaches with an encyclopaedic knowledge, although humans contribute meaningfully by drawing connections and a…

> I'm highly bullish on their capabilities as a force multiplier, but highly bearish on them becoming self-driving (for anything complex at least). Very well summed and this is my exact stance, it's just that I am not seeing much of the "force multiplier" thing just yet. Happy to be proven wrong, but last time I checked (August 2024) I didn't get almost anything. Might be related to the fact that I don't do throwaway…

Do not know how you are using them? It speeded up my development around 8-10x, things i wouldnt have done earlier i'm doing now, e let it do by the AI; writing boilerplate etc. Just great!

Re: Ask HN: SWEs how do you future-proof your career in light of LLMs?

#970

Post by pdimitar »

ta12653421 wrote: post #969
pdimitar wrote: post #821

Earlier quoted context omitted.

> I'm highly bullish on their capabilities as a force multiplier, but highly bearish on them becoming self-driving (for anything complex at least). Very well summed and this is my exact stance, it's just that I am not seeing much of the "force multiplier" thing just yet. Happy to be proven wrong, but last time I checked (August 2024) I didn't get almost anything. Might be related to the fact that I don't do throwaway…

Do not know how you are using them? It speeded up my development around 8-10x, things i wouldnt have done earlier i'm doing now, e let it do by the AI; writing boilerplate etc. Just great!

8-10x?! That's quite weird if you ask me.

I just used Claude recently. Helped me with an obscure library and with the hell that is JWT + OAuth through the big vendors. Definitely saved me a few hours and I am grateful, but those cases are rare.

Post reply on HN