Live data from Hacker News

2x, not 10x: coding with LLMs in 2026

obryant.dev

251–260 of 260 posts

Re: 2x, not 10x: coding with LLMs in 2026

#251
post #244

Earlier quoted context omitted.

> AI is making me work more (good thing) How is that good?

It makes me feel fulfilled. It gives me things I want. Building is in my nature. The work I'm doing is for myself (I'm not GP). But work isn't just "my job for the man." You can build things for your home, your shop, your family, your business if you have one.

Then call it your hobby, right?

Re: 2x, not 10x: coding with LLMs in 2026

#252
post #237
post #192

Earlier quoted context omitted.

Claude lies, like all LLMs today.

Oh, right. Let's throw it all in the trash, totally useless

Why would you say that? It's incredibly useful technology that also lies. It's not binary. There are tradeoffs. Just be careful.

Re: 2x, not 10x: coding with LLMs in 2026

#253
post #211

Earlier quoted context omitted.

We’re at the point where people who talk too fancy get accused of being LLMs, and the people who make human mistakes get scolded for grammar.

It seems like constructive criticism rather than scolding. If I were running around saying "providence" instead of "provenance" I would want someone to tell me.

Yes, it was meant constructively and for exactly this reason. I interpreted what they wrote as what they meant, then realized it was a mistake that I could have easily made too.

Re: 2x, not 10x: coding with LLMs in 2026

#254
post #54
post #51

I'm not sure about the x, but the first thing that arises from that is, I feel like in my case it's way higher than 2. The 2nd thing is, how do I measure that. --- In my case, the details of my work (Kinda DevOps, kinda Senior Dev) makes it that having an LLM to do the heavy lifting allows me to do things not only faster, but better, and across domains I do not hold expertise on. An example of the effect of LLMs in m…

> allows me to do things not only faster, but better, and across domains I do not hold expertise on. How would you know it's better when you have no expertise?

[flagged]

Re: 2x, not 10x: coding with LLMs in 2026

#255
post #252
post #237

Earlier quoted context omitted.

Oh, right. Let's throw it all in the trash, totally useless

Why would you say that? It's incredibly useful technology that also lies. It's not binary. There are tradeoffs. Just be careful.

My point is that far too many people that offload everything to AI tend to just accept the text it generates without thought or verifying what it generates.

Re: 2x, not 10x: coding with LLMs in 2026

#256

Earlier quoted context omitted.

> to do this I had to give up reading or even controlling the code and focusing on behavior/design-level control (not superficial, still dictating overall technical architecture) how do you verify the behavior? are you still writing or at least reading tests or just doing manual testing?

I discuss the testing approach and coverage with the model before and after, sometimes in a fresh thread that does a static analysis. interestingly my input is still pretty important

> interestingly my input is still pretty important

how do you know?

Re: 2x, not 10x: coding with LLMs in 2026

#257

Earlier quoted context omitted.

> gave me a checklist that helped me quell my travel anxiety. it's truly fascinating how many positive descriptions of AI gesture at emotional management. I think that's the killer feature of this technology -- it makes people feel good, capable, reassured -- without the risk and vulnerability of interacting with another human. I use Google Weather for my forecasts btw, no need to vibecode an app for that

Gathering forecasts for each leg of a trip only on the particular days for the particular location is tedious. Anyone can do it with google weather but why spend the time when you don't have to. I can keep track of my expenses on a napkin but i'd much rather use a spreadsheet or dedicated app especially when that app is effectively free.

did you know that Google Weather has saved locations?

Re: 2x, not 10x: coding with LLMs in 2026

#258
post #238
post #106

I recently removed about 50% of code in a feature submitted by a colleague, because it's a pile of over-engineered mess that either never gets used or caused trouble for us. We probably would not have added the code in the first place if we actually discussed the design. With the code removed, the feature is now much simpler and more maintenable. And that's the second time this happened over the past few months.

One time before LLMs we outsourced this project. It was something to sign file headers, written in ruby by an overseas company. It was something like 4500 lines long, and I wound up reducing it to something like 550 lines. My story, like yours, has nothing to do with the topic at hand. It happened many times to me, as well. Humans write terrible code. You didn't even mention if the code you're talking about was gener…

I think my last paragraph should have provided enough context.

Re: 2x, not 10x: coding with LLMs in 2026

#259
post #47
post #26

You still have to know what you’re doing. That came from years of doing it by hand. Where is that necessary experience going to come from for the current and future generations of juniors?

Shipping. Breaking stuff. Fixing it. And discovering, repeatedly, that the tool sounded much more confident than it had any right to. Like always. This is why the average carpenter you hire in 2026 still won’t do a perfect job, despite carpentry having existed for thousands of years.

So you don't think fundamentals built by actually coding will be necessary to produce well structured and maintainable code with an LLM?

Re: 2x, not 10x: coding with LLMs in 2026

#260
When you look outside of coding, the productivity boost and time saving also hits when you're using the Ai as a thinking partner. eg there is a significant different between Opus and Fable when it comes to thinking, data analysis, strategy. I also think Grok 4.5 is extremely underrated, faster than any other model, less verbose, reliable. The human to agent trust factor is also under-considered - how much does the agent "get" you when you give it less.
Post reply on HN