Ask HN: Is software engineering just patching APIs and wrangling Kubernetes?

I think you'd find other physical engineering jobs are very similar at scale.

Most poeple out there building new rail are building them in situations similar to how other people have already built rail: apply known best practices given the intricacies of the current situation. Very few people are doing things like "building the worlds first rail line across a floating bridge" (Seattle I-405).

Even something like DeepSeek in China, wasn't necessarily so much about big AI breakthroughs as it was low level optimization of existing hardware. At the end of the day, even big achievements are built on piles and piles of what seems like drudgery, I guess it's just about feeling like that drudgery is contributing to something great.

I do the first two.

I've worked for orgs that thought about Kubernetes but not any that pulled the trigger. Let's face it, we used to run web sites that had millions of users out of a single instance of mysql 20 years ago. You could fit all the people who admin web sites that really need distributed systems in a big conference hall in Las Vegas, if you deleted Google you could fit them all in the Moscone Center. (SV companies wouldn't send a bunch of kids to Vegas though because they wouldn't trust not to get in them w/ the alcohol, the cards and the hookers)

I'm not impressed that anybody got something done in two years with 100 engineers that got paid $300,000 per year and aren't saving a dime because they're paying it all to landlords who were far sighted enough to buy property around the San Francisco bay before there was a Google. I am impressed when 2 or 3 people that could have starred in a Wix commercial and live in a flyover state get something done over the weekend.

I love being able to make a smart RSS reader, an image sorter that works with my tablet and my VR headset, web applications for car dealers and vineyards, social networks for secret societies and a great web site for searching public opinion data that makes our competitors look like a bad dream. Sure it took me a while to stop worrying and love React but I think it's a great time to be alive.

The role has absolutely been degraded. The SWE role has been distorted by the industry. The responsibilities have been eroded and all the potential specializations of it have been bundled together, including responsibilities that have nothing to do with engineering, and even so most companies will insist this is this "engineering" and demand that you adhere to it.

Ironically, the review process does not reflect any of this. You are asked technical questions as if you really are going to do any of that on a day to day basis, or even have that kind of mindset. The usual attitude that is expected has more to do with business and making profit rather than engineering good systems.

> rather than building something truly exciting.

What were we doing that was so exciting a decade ago? We were stitching together different APIs, wrestling with jQuery, and managing servers.

I mean, you're not wrong - someone was out there creating React, and Kubernetes, and AWS, and programming languages, all the stuff we're using now. But most of us were still doing the same stuff. And turns out, it still paid the same bills.

The #1 skill of a software developer is to translate incomplete specs from other humans into a product.

Get into a different part of the industry: I’ve never touched any of that stuff and have gotten on quite fine for 30 years so far. From the outside, the software engineering world of the web-centric tech industry is one that sounds absolutely dreadful (and has for a while). There is a ton of software being written that isn’t nearly as miserable.

Of course, if ones metric is “FAANG or nothing” or maximizing compensation, you get what you seek - which may be miserable work in exchange for status and cash. I realized a long time ago that neither status nor cash outweighs enjoying what I do.

it's funny how we keep solving the same problems over and over. While we have these great frameworks why is it so hard to make a product that has desktop, web, mobile and tablet options. If anything development seems slower and more costly now then 10 years ago.

I mean, most of the "innovation" in the software world stopped when smartphones became a thing, and the target shifted towards monetizing those things. Some things are happening in AI, but again its all about monetization rather than pushing technology.

There are some cool software jobs that deal with embedded SoCs, but those don't pay nearly as well.

I was feeling a bit this way, but ever since going independent and working on my own startups/open source projects, I don't feel that way. It's a lot more stressful and inconsistent, but the work is more interesting. I think modern AI tools help with a lot of the tedium of gluing APIs together and things like that.

1. It has bever been an art. It had always been just a mean to an end, a job.

2. If you’re in a big company that is the part of the game. By design. Want to xcitement - join a startup.

Sounds like a problem provincial to the domain you're working. Sounds like enterprise microservices and webapps?

You've described my job, but I've found some excitement in the "distributed systems" part of it. It's not trivial to make the right design decisions to make a microservices architecture run well. If it was trivial, we wouldn't have so many problems at work.

That said, I do kind of miss the programming itself being interesting. I did more difficult programming at university.

Never look to your job for fulfillment in the art. There are very few situations where you can create art at your job. That’s what our free time is for, if you care enough to.

You're not wrong

In my experience if you want interesting work you need to seek out a more niche domain like systems programming, graphics, computer vision, audio, embedded, AI, GPU programming, robotics, gamedev, etc...

Something that hasn't been streamlined/optimized to death by 1000 other developers

Otherwise its just wiring together APIs, shuffling data around, and wrangling deployments/infra.