The complexity stuff is all absolutely true.

However I think it's aggrandizing what human engineers actually do with remarks like "Engineers own tradeoffs." My experience is that certainly less than half of the employed software engineers don't actually give a real analysis to questions like:

"Given these constraints, this team, this business, this infrastructure, this budget, these risks, and the expected evolution of the product, what is the most appropriate way to implement X, today?"

Thus I think AI is more able to replace the average engineer more than this article admits, however the inadequacy of "average engineering" will be much more apparent now: codebases can become large/complex enough to be unwieldy in months now when it used to take 5 years [a timescale where accountability is effectively impossible].

▲

KronisLV a day ago | parent | next [-]

> this team, this business

These get overlooked so often. The way you build software if you’re at the helm vs the way you need to build it when dealing with a more/less capable team and business, especially if someone else will be doing the deployment and will need lots of consultations, is way different.

	▲	hn_go_brrrrr a day ago \| parent [-]
		This is my favorite part of software engineering. It's not just a set of rules you can apply to get the right answer. You need to use your judgement to make a context-appropriate decision.

▲

therealdrag0 21 hours ago | parent | prev | next [-]

Yep that’s all “senior+ engineer” stuff at least and even those can be cut in half by quality. So maybe 25% of engineers have those attributes?

▲

jaen 18 hours ago | parent [-]

I think you overestimate the amount of senior engineers... even at the extreme end of top-tier big tech, only around 30% can be considered senior. Staff engineers who regularly think at the level GP talks about are 5-10%.

Everywhere else the percentages are quite a bit lower still (if judged by the same standards).

	▲	therealdrag0 17 hours ago \| parent [-]
		Ya I was very hand wavy. Though in my experience a lot of smaller shops have a higher percent of seniors than big tech. And I don’t know how much smaller shops there are, so I rounded up. Also smaller shops depend more on this well rounded thinking.

▲

lovich 15 hours ago | parent | prev | next [-]

Do most companies even allow that kind of engineering to occur?

Prefacing that I’ve never worked at a faang. At more than half the companies I’ve worked out most of those decisions were made by non technical leadership for non technical reasons. Ranging from the reasonable(our predecessors signed a deal with Oracle a decade ago and violating it will cost us more than this project is worth) to the unreasonable(I had lunch paid for by this vendor so we’re using them now).

I also frequently ran into the problem of the process of doing that level of engineering requiring the business to make choices and being completely incapable of it. I could give 3-4 different plans with explicit tradeoffs, both in detail and with an executive summary, and what they meant for the company and even that low number of choices induced analysis paralysis in the management but they barred me from doing anything until they made the call.

▲

echelon a day ago | parent | prev [-]

It's been ten months since good models started landing and threatening our current job descriptions.

Do you think this is where it stops? This is where it begins.

Machines will be good at managing complexity too. You can't draw a line and say improvement stops here, because everything we've seen so far flies in the face of that.

I shudder to think what these models will be capable of in 24 months.

▲

preommr 21 hours ago | parent | next [-]

> Do you think this is where it stops? This is where it begins.

No, this is pretty much where it stops.

The models are good enough for the average coding task, and the slop they produce often is in the category of what a bad or careless dev that's being contracted out might produce.

Yea, they'll get better, but not in next-level sort of way.

The limitation is not the models or intelligence, it's the human in the loop. We're still stuck on stupid human issues ranging from usability bugs, to figuring out what the product should be, to how we should program in the first place.

I know the models are capable of sorting out issues it gets stuck on because it's writing error handling in the wrong way, or just it doesn't have the right abstractions, because we can't settle on the right way to program. I still see people arguing about dyanmic vs. static typing.

And obviously, there is a next level, but that's real singularity, and we're all out of jobs.

▲

therealdrag0 21 hours ago | parent [-]

I think the limitation is still context. Humans are better at judgment than AI because humans have more context to put together better decisions. when AI has all of the context, It beats the socks off me.. for better or worse sharing and organizing context is a notoriously difficult problem that won’t be solved overnight. This could easily be a 10 or 20 year problem because it depends on human organizational structures and investing in free flow of information and information storage tools. Unifying search tools like Glean help a lot here but they’re also expensive.

	▲	dwaltrip 13 hours ago \| parent [-]
		100% agreed. I consider Fable to be a super genius that was born yesterday and has memorized the internet, yet barely understands humans on a deep behavioral level. By the end of a session, sometimes Fable and I are cooking incredibly, but then alas we have to start a fresh session. A lot of the intangibles about what are going well at that point are extremely difficult, if not impossible, to capture compactly in a “handoff” or “guidelines”. The lived experience of working with me through the session, as represented in the current context, is what produces the higher quality results. And likewise, the naivety it has as a newborn at the start of every session, along with the lack of deep human behavioral understanding, explains why these cutting edge models like Fable can still be so dumb in some ways while being mind-blowing in other ways.

▲

HarHarVeryFunny 19 hours ago | parent | prev [-]

> Machines will be good at managing complexity too.

Sure - eventually.

The question is where is the training data going to come from.

Without "training data" (the non-existent diaries of people designing complex systems and recording their thought processes) you're stuck where we are today, where the LLM is basically doing cargo-cult design and decision making - copying the outcome of decision making (problem looks like X, so I'll use design pattern Y, same as most people do), without understanding why those decisions were made.

How much of the time this matters remains to be seen, as people try to use LLMs to help design more complex bespoke software, rather than just yet another CRUD app or three.js game.