Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jf22
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
jf22
3mo ago
I like AI tools. I'm bullish on AI tools. I'm productive with AI tools. But... At my job the prevailing sentiment is everything should be easy and fast with AI now so why bother doing anything that can't be done with AI. Talk
2.
▲
by
jf22
4mo ago
But we aren't cooking with gas. We are cooking with a more controlled burner than ever that can download a clean code claude skill and be committing better code than you or I could write. What would normally be considered overengineere
3.
▲
by
jf22
4mo ago
First, most software is already a hot mess. Second, LLM code can be less of a hot mess than human written code if you put in the time to train/prompt/verify/review. Generating perfect well patterned SOLID and unit tested code
4.
▲
by
jf22
4mo ago
What do you mean?
5.
▲
by
jf22
4mo ago
What's a good example of human-led design?
6.
▲
by
jf22
5mo ago
I'm interested to know why you think data structures are important. AI is pretty good at reasoning out data structures problems.
7.
▲
by
jf22
6mo ago
If there isn't a phrase then just say the numbers...
8.
▲
by
jf22
6mo ago
You can tax reach though.
9.
▲
by
jf22
6mo ago
The "overwhelming majority" standard for harm seems odd when you use 15% of smokers getting harmed as an example. 15% is not an overwhelming majority.
10.
▲
by
jf22
6mo ago
We are way past this question now.
11.
▲
by
jf22
6mo ago
I kept a personal license for while because I thought Windsurf kept memories a little better than Claude. But now since I use Claude at work I switched to keep my home and work stacks similar.
12.
▲
by
jf22
6mo ago
All that cleanup can be done with AI and then you can add guidance to the codebase so that the next LLM will do a better job at conforming.
13.
▲
by
jf22
6mo ago
We are not.
14.
▲
by
jf22
6mo ago
I feel like I'm a stone skipping across the water instead of being submerged deep in coding context. I'm getting used to it but I don't like it. I'm also much more tired at the end of the day. My mind feels broken. I don
15.
▲
by
jf22
7mo ago
How are the architecture changes you are proposing improving the end result? >but not being able to control the exact output worries me Why?
16.
▲
by
jf22
7mo ago
I have a physical goods side hustle already and I'm brainstorming ideas about a trade I can do that will benefit from my programming experience. I'm thinking HVAC or painting lines in parking lots. HVAC because I can program smart
17.
▲
by
jf22
7mo ago
I don't understand he comment. AI produces code that people can understand and is easy to maintain if you ask it for that.
18.
▲
by
jf22
7mo ago
I'm an 99% organic person (I suppose I have tooth fillings) and the new models write code better than I do. I've been using LLMS for 14+ months now and they've exceeded my expectations.
19.
▲
by
jf22
7mo ago
I'm a former EM who would never go back in an AI age. EMs deal with friction and from my experience more output is more friction. You have org leaders and businessy people putting their foot on the gas because AI is so productive and t
20.
▲
by
jf22
7mo ago
This is a really bad take (and bad faith) because all the failed tech initiatives you mentioned barely had any adoption whereas LLM based AI tools are used by a billion people a week.
21.
▲
by
jf22
7mo ago
I think the question is why didn't you ask the AI and then if you need more information ask the human including what you learned from AI.
22.
▲
by
jf22
8mo ago
Humans, with reliability, will ignore the patterns or best practices in your code and implement the same thing twice in different ways. In 25 years I've never worked in a codebase where this wasn't true.
23.
▲
by
jf22
8mo ago
Sure but in my experience LLMs behave much more consistently than humans with regards to quality. LLMs don't skip tests because they have to make a deadline for example. And now we have LLMs that review LLM generated code so it's
24.
▲
by
jf22
8mo ago
You can tell LLMs to conform to certain quality standards and code in a way you think is best. Gold plating code with the best quality standards takes LLMs seconds whereas it would take you days doing it by hand.
25.
▲
by
jf22
8mo ago
The current AI tools extremely good at telling you about how an existing system works. You don't need that one super knowledgeable person anymore. With the right MCPs or additional context you can have the tools go read PRs or the last
26.
▲
by
jf22
8mo ago
Because it's such a weasel "just asking questions" thing to do. If you had a concern about the material costs of renewables you should know what they are and if you wanted to have a good faith discussion, you'd also be a
27.
▲
by
jf22
8mo ago
I'd disagree the fervor is religious. I think it's more frustration. Pointing out there is a maintenance cost to infrastructure is silly and doesn't add to the discussion. We all know materials have to be shaped into machines
28.
▲
by
jf22
8mo ago
I'm not sure what you're trying to say here. Canada can be just as corrupt.
29.
▲
by
jf22
8mo ago
Do you agree or disagree you provided a link to a fitness influencer's website as "evidence?"
30.
▲
by
jf22
8mo ago
Yes the workflow has shifted. I've handled the sloppiest slop with llms and turned the worst code into error free modern and tested code in a fraction of the time it used to take me. People aren't worried about cost because $1k in
More ›