jascha_eng

6 ms·

61.
by jascha_eng 8mo ago
@dang this accounts comments smell like LLM slop. They are mostly on topic and its more claude than chatgpt but it's slop nontheless. is telling didn't win... It won ... Look at their other comments they are also fishy I know you
62.
by jascha_eng 8mo ago
That's how you slow down development to a crawl
63.
by jascha_eng 8mo ago
One solution is to get rid of anonymity online, enforce validation of identity. Every human only gets 1 account. And then we still ban people that use AI. Might take a bit but eventually we'll have filtered out all the grifters. If tha
64.
by jascha_eng 8mo ago
I noticed this on my spring boot side project. Successful test runs produce thousands of log lines in default mode because I like to e.g. log every executed SQL statement during development. It gives me visibility into what my orm is actual
65.
by jascha_eng 8mo ago
> the "intelligence" was never in the input It's quite literally in the authors prompt so in the input. it's in the article that without his prompt the gibberish input produces nothing of value: "Hello! I am an
66.
by jascha_eng 8mo ago
This actually used to be one of my favorite interview questions for backend engineers. Everyone has used transactions but depending on your seniority you'd understand it to different degrees. And no I'd never expect people to know
67.
by jascha_eng 8mo ago
> well-aligned developers I think this is very key, if the development style and the direction of the project is clear, much less review and alignment is necessary. And also > avoid keeping branches open for more than a day Big +1 on
68.
by jascha_eng 8mo ago
I've rarely seen the first description of it where people actually commit directly to main. Except in very early stage projects. But it does always feel the fastest if you only review code "on-demand" in PRs/MRs instead
69.
by jascha_eng 8mo ago
lmao I love how stupid that response is.
70.
by jascha_eng 8mo ago
most importantly it is watermarked as ai created. Thank you
71.
by jascha_eng 8mo ago
idk I haven't really hit the point with any llm that it comes up with useful abstractions on its own unless those abstractions have been in the training data. E.g. imagine building a google docs clone where you have different formattin
72.
by jascha_eng 8mo ago
My impression so far is that the parallel agent story is a fabrication of "ai influencers" and the labs themselves. I might run 3-4 claude sessions because that's the only way to have "multiple chats" to e.g. ask un
73.
by jascha_eng 8mo ago
> OpenClaw is more like an art project than a consumer product I think this is true for a lot of vibe coded applications. Never thought about calling it an art project but it hits home.
74.
by jascha_eng 8mo ago
Yes and if was done with people using cursor at the time and already had a few caveats back then about who was actually experienced with the tool etc. Still an interesting observation. It was also on brown field open source projects which i
75.
by jascha_eng 8mo ago
I think people got fatigued by reviewing already. Most code is correct that AI produces so you end up checking out eventually. A lot of the time the issue isn't actually the code itself but larger architectural patterns. But realizing
76.
by jascha_eng 8mo ago
Why? They had $10+ billion arr run rate in 2025 trippeled from 2024 I mean 30x is a lot but also not insane at that growth rate right?
77.
by jascha_eng 8mo ago
Also GPT´5 when I ask: > I want to wash my car and the car wash is only 100m away. Do you think I should drive or walk? It responds: Since it’s only 100 meters away (about a 1-minute walk), I’d suggest walking — unless there’s a specific
78.
by jascha_eng 8mo ago
There are a lot of non developer claude code users these days. The hype about vibe coding lets everyone think they can now be an engineer. Problem is if anthropic caters to that crowd the devs that are using it to do somewhat serious engine
79.
by jascha_eng 8mo ago
This is a great take. It applies the SaaS is dead theory at a lower level (libraries are dead) but has it a much more nuanced view. Yeh even if LLMs are 10x better than today you probably still don't want to implement cryptography from
80.
by jascha_eng 8mo ago
verbatim llm output with little substance to it. HN mods don't want us to be negative but if this is what we have to take serious these days it is hard to say anything else. I guess I could not comment at all but that feels like just l
81.
by jascha_eng 8mo ago
The hard part of leet code is not the coding but learning to think about problems the correct way. You can solve leet code problems on the white board with some sketches it has nothing to do with the code itself.
82.
by jascha_eng 8mo ago
Noone can correctly quantify what these models can and can't do. That leads to the people in charge completely overselling them (automating all white collar jobs, doing all software engineering, etc) and the people threatened by those
83.
by jascha_eng 8mo ago
All of these skills always having custom installers to place a few markdown files is such a mess.
84.
by jascha_eng 8mo ago
I ran the command, looked at the report. It said something like 300+ sessions are about database migrations. That can't be its a new project I only had like 10 claude code sessions so far. I asked Claude about it it looked into some fi
85.
by jascha_eng 8mo ago
You just don't know which parts of the doc are real and which are hallucinated. Maybe the prompter checked everything and the content is actually good, but sadly many don't and there is a lot of slop flying around. So if you want
86.
by jascha_eng 8mo ago
I've had this with multiple datadog libraries. Even offered to implement some features for the python one for free because we were missing a feature but they just never responded. Note that they had this feature in their php lib so it
87.
by jascha_eng 9mo ago
This does not normalize for tokens used if their skill description was as large as the docs index and contained all the reasons the LLM might want to use the skill, it likely performs much better than just one sentence as well.
88.
by jascha_eng 9mo ago
Fair but if spam becomes an issue that blocks good research from happening maybe adding some filters improves the end result more.
89.
by jascha_eng 9mo ago
Why not filter out papers from people without credentials? And also publicly call them out and register them somewhere, so that their submission rights can be revoked by other journals and conferences after "vibe writing". These a
90.
ANN v3: 200ms p99 query latency over 100B vectors(turbopuffer.com)
2 points by jascha_eng 9mo ago|0 comments

More ›