Top
Best
New

Posted by poisonfountain 7 hours ago

LLMs are eroding my software engineering career and I don't know what to do(human-in-the-loop.bearblog.dev)
680 points | 633 commentspage 3
wcfrobert 2 hours ago|
> "Now I have CLIs that one-shots bugs across distributed systems for me. Bugs that I couldn't solve in the past. Bugs that would take 2 days of full-time debugging. Bugs across distributed systems that lack distributed observability. 90% of the bugs are one-shotted now, including bizarre race conditions, unexpected corner-cases, third-party integration issues, undocumented API edge cases, everything. I hardly have to intervene."

The fact that the author can articulate _why_ the AI is getting so good is kind of a moat for specialist, right? Imagine a layman prompting without domain expertise:

"There is likely a race condition here + [long-winded explanation and analysis carefully guiding the AI]"

Degenerates to:

"This button is not working, please fix. I don't care about code. Decide yourself"

Degenerates to:

"Claude make me money"

Aperocky 4 hours ago||
> And then I started realizing: all the knowledge I have accumulated over the years: the trade-offs between implementations, how acquiring works, how to structure idempotency to prevent double-charges, everything, was becoming useless.

How is that true? I've been using Opus on an industry scale over last 6 months and this is just not real.

It has consistently with a certain percentage of chance each time (and no claude.md and skills do not stop it fully):

* Suggested to remove tests to allow for things to pass

* Suggested remove an error so that things can be "unblocked"

* Suggested to use a second path when the original path ran into problem instead of making the original path accomodate for that possibility.

* Suggested or silently added "features" or "guardrail" that I don't want.

* Can be left unsupervised only if given a goal that it can verify against itself. Without such clear goal (e.g. this test in the integration environment must be fixed), it flounders.

I'm not using just the native harness (e.g. CC) either, with additional, customized harness, the behavior improves somewhat but are still fundamentally constrained and cannot really be trusted without verification.

See my methodology (100% handwritten): https://aperocky.com/blog/post.html?slug=agentic-development....

Being a heavy user I think I've ran into every single hallucination that the model can do over development release and operations. I am still a heavy user but there are a lot of value in recognizing where exactly LLM's limit is and work around that.

causal 4 hours ago|
To me the greatest monument to Claude's poor software quality is Claude Code itself.
Aperocky 4 hours ago||
Yes, let's build a 40K line main loop! I wonder if they thought claude code need to be more like an LLM to work lmao.
PeterStuer 4 hours ago||
I think the one thing the author is underestimating, especially in his "first pillar" is that he is able to coach the models into great results because he already knows the lay of the land. I often make the same mistake. I see people struggle with GenAI, and feel flabbergasted how they succeed to fail, but then if you observe how they work the tool, it is clear they have no idea what to ask or how to evaluate and iterate on the responses.
rglover 2 hours ago||
Based on what I've been seeing/reading online the past ~6 months or so, I think there's a self-fulfilling prophecy going on here.

Developers are concerned about jobs going away, but how often are they pushing back in their orgs about how AI works? In response to "are you using AI to move faster," how many are responding with "yes, but there are some things you should know..."?

If there's no pushback and just pure acceptance of stuff like tokenmaxxing, then what does anybody expect when the broader narrative around AI is that it can help a novice to grind out miracles (i.e., "holy crap, if this is what a novice can do, what can an expert do?!")?

Of course leadership is confused because (it seems) few are asserting expertise, saying "no," and stating a clear case as to why they're doing that.

The default excuse is "I don't want to lose my job" (which is a fair reaction to all of this, especially these days), but it's worth considering when/how that choice is actually just shooting future you in the foot later. It seems there's a broader trend toward compliance more than there is "you hired me to do this job properly, did you not?"

rybosworld 1 hour ago|
For most companies that I've worked for, pushback is normal and expected as long as you've built some trust/rapport with your management chain.

However with AI, it feels different. I have seen both technical and non-technical managers tell engineers something to the effect of "you aren't prompting correctly" if they aren't able to get the task done within some preferred time frame.

We are seeing the industry revive metrics like lines of code, number of tickets closed, bug's found (looking at you Mythos), and now even "tokenmaxxing". It's exhausting to push back on. These are all things that we know will be gamed. But the individual that brings this up might be viewed as "anti-ai" or something.

If you're an IC, I do think the best thing to do is just go along with it. Sooner or later we will see more shocked-pikachu-faced executives when they realize that engineers are spending tokens just for the sake of it.

bluefirebrand 9 minutes ago||
> If you're an IC, I do think the best thing to do is just go along with it

I personally think the best thing to do is start retraining now so you aren't screwed by the time this all topples

PLenz 1 hour ago||
Start advocating for a wealth tax
jchw 7 hours ago||
Same boat here, just a couple years more experience.

Current LLMs are still kind of shit at actually programming so many jobs do still care to have professional programmers. However, I think it's evident that if things stand where they are, employers will care to have far fewer of them, at least of highly paid highly experienced programmers. If this is the state we're in with LLM adoption when they can't help but create the same helper functions 15 times, god knows we're screwed.

So we should probably work on clearing out our debts and figuring out what else we might want to do with our time, I reckon.

I'm still going to try to do a good job. I'm still trying to learn the best effective ways to apply current LLMs (Right now I still prefer to mostly write code myself but have been using LLMs to bang code into shape via iterative code review; this is a way to exploit LLMs to make better code, especially applicable if your velocity was already good.)

hypfer 6 hours ago||
This feels fake/engineered but regardless of that also redundant.

It's the exact same story that we've heard countless times by now. Hosted on a blog with just a single post. Named in a way that suggests that said blog was created for this very single post.

What is there to learn from this other than LLMs seem to be bad for some people's psyches and that AI companies need these very stories to not get their funding shut down?

Havoc 3 hours ago||
>Hosted on a blog with just a single post

Would you put a "Hey i'm feeling a little useless" post on your main blog / linkedin?

gamegod 5 hours ago||
It's 100% fake, and half the comments here are from 20 year olds working at AI companies.
hypfer 5 hours ago||
I mean I kinda get that it sucks being a junior now, but otoh, it might also not?

It might be easier to adapt to this new tech when you're 19 compared to when you're 59.

But honestly, this discussion _also_ has happened ad-nauseam by now. Everything that was worth saying has been said. And then some.

People don't actually want to talk about LLMs. They want a hug. And that's fine, human and all.

But could you please just start asking for hugs instead of encoding that into vaguely profound sounding takes on AI? I'm tired of this play pretend.

dwh452 5 hours ago||
There are lots of positives that have resulted from using AI in software engineering. (1) No more long repetitive text editing sessions. I.e. changing namespaces or replacing deprecated APIs with the "correct" ones. AI will make nearly perfect text modifications with ease. (2) No more bike-shedding code reviewers nitpicking over every tiny coding decision. I.e, "you should use std::format instead of std::stringstream". AI will match the existing set of nitpicks so you don't have to. (3) Average Joe's and Jane's can craft applications by just talking to the computer. This might inject a freshness to the current state of software. Currently, we are all forced to use the same bloated applications like Word, Excel, Jira, and Photoshop. We are currently forced to deal with the same set of monopolistic SW companies. Now average folk can solve problems and avoid dealing with Microsoft for a spreadsheet program.
manishsharan 5 hours ago|
Average folks like their Excel and Word. Most families have MS subscriptions like they have Netflix subscriptions.

Monopolies will continue as Token prices continue to rise.

gaiagraphia 6 hours ago||
There's a certain irony in masters of automation lamenting that their roles are being automated. I wonder whether the jobs their efforts eroded in the past ever got the same thoughts...

Programming, logic, etc are skills and toolkits. The optimal state of society is everybody being able to apply them, not just the enlightened compsci caste. There was a time in the past where scribes were paid nice cash for their efforts, too.

I guess the lesson to learn here is treating a toolkit as an identity and job for life. By virturee of the essence of the job itself - if the tool gets cheaper and more widespread, it's aactually success, not betrayal.

tines 6 hours ago|
You say that the optimal state of society is for everyone to apply programming and logic etc. but the obvious final result of these developments is that no one will.
gaiagraphia 6 hours ago||
Maybe the artform will be lost, but surely humanity will inherently be more 'logical' and systems driven afterwards?

Maybe using writing as an analogy is flawed, but most of humanity having 'writing' as a core skill did enable many other things, even if oral storytelling cultures suffered at its hand.

At its core, tech is all about breaking through inefficiencies and barriers. Does it matter if people can't code python if people demand government systems be frictionless in the year 2500?

jplusequalt 5 hours ago||
Sincerely, how is prompting an AI to build software for you building "logic and systems thinking"?

The thing many people are ringing the alarms over is the offloading of critical thinking and knowledge work to LLMs.

gaiagraphia 4 hours ago||
Being able to program isn't the end game of critical thinking. Programming languages are just a way of representing the processes. The thinking underneath was always more important, and there's now technically more time freed up to focus on that. Billions of people now have access to tools which will aid them in reasoning through complex problems without needing a $100k CS degree. Of course some people are using LLMs to get recipe inspiration, but others are now empowered to do things which were impossible for them before.

I personally think the alarm ringers are mainly the privileged elite who are scared of their moats beyond filled in. LLMs have effectively broken down the gates of access to knowledge. In a diverse world, having more people being empowered to do more things has to be a net positive.

tines 3 hours ago||
You clearly haven’t observed anyone actually using AI.
gaiagraphia 2 hours ago||
I've come across lots of novel uses, though. I've seen non-technical people in law, education, hr, all spin up interesting projects/workflows which have helped their jobs/lives. The most interesting was a teacher who's dropped powerpoint, and is creating interactive lesson experiences; mainly presenting the content through popular games and creating engaging activities to apply the content. A recent one was a mock UN summit presented like a game of Civ with territory mechanics. Absolutely zero tech background, just curiosity.

Once people get over a few hurdles, things like: >tech's too confusing >$20 is a lot of money to spend on a subscription >AI is just a fancy search engine >AI will do all the work for me

You start unlocking a fair bit of creativity in people. I mean, all this is brand new stuff even for tech-savvy people. It'll take a while for the genuinely useful uses to dissipate out into the maasses.

Not everything has to be a billion dollar business.

mullenba 6 hours ago|
I've consulted with some big companies on AI strategy. I tell them there are two approaches to AI.

1) Train AI to replace human work. This gives you 50% quality for 10% cost. 2) Train AI to assist human workers. This gives you 200% quality for 110% cost.

Most companies will go with option 1, and it's a race to the bottom. Eventually, someone will go with option 2 and gather up all of the pieces and take over the market.

i5heu 1 hour ago|
This is not at all how all of this works.

If you train an AI in one thing it will become better in the other.

More comments...