orange_puff

I basically fully agree with this. I am not sure how to handle the ramifications of this in my day to day work yet. But at least one habit I have been forming is sometimes I find that even though the cost of writing code is immensely cheap, reviewing and validating that it works in certain code bases (like the millions of line mono repo I work in at my job) is extremely high. I try to think through, and improve, our testability such that a few hundred line of code change that modifies the DB really can be a couple of hours of work.

Also, I do want to note that these little "Here is how I see the world of SWE given current model capabilities and tooling" posts are MUCH appreciated, given how much you follow the landscape. When a major hype wave is happening and I feel like I am getting drowned on twitter, I tend to wonder "What would Simon say about this?"

This is also my feeling. When people keep referring to big jumps or inflection points, I am left confused, because the models have felt good for a long time and feel like they are getting steadily better. This could be biased by what I use them for though.

Do you mind elaborating? By API design, do you mean how they structured their classes, methods, etc. or something else?

This seems really impressive. I am too lazy to replicate this, but I do wonder how important the test suite is for a a port that likely uses straight forward, dependency free python code https://github.com/EmilStenstrom/justhtml/tree/main/src/just...

It is enormously useful for the author to know that the code works, but my intuition is if you asked an agent to port files slowly, forming its own plan, making commits every feature, it would still get reasonably close, if not there.

Basically, I am guessing that this impressive output could have been achieved based on how good models are these days with large amounts of input tokens, without running the code against tests.

I used to think this way too. Here are a few ways I've tried to re frame things that has helped.

1. When I work on side projects and use AI, sometimes I wonder "what's the point if I am just copy / pasting code? I am not learning anything" but what I have come to realize is building apps with AI assistance is the skill that I am learning, rather than writing code per se as it was a few years ago.

2. I work in high scale distributed computing, so I am still presented with ample opportunities to get very low level, which I love. I am not sure how much I care about writing code per se anymore. Working with AI still is tinkering, it has not changed that much for me. It is quite different, but the underlying fun parts are still present.

I have been trying to find such an article for so long, thank you! I think a common reaction to Agents is “well, it probably cannot solve a really complex problem very well”. But to me, that isn’t the point of an agent. LLMs function really well with a lot of context, and agent allows the LLM to discover more context and improve its ability to answer questions.

As others have mentioned please add more docs / details to the README

I want to mention my current frustration with cursor recently and why I would love an OSS alternative that gives me control; I feel cursor has dumped agentic capabilities everywhere, regardless of whether the user wants it or not. When I use the Ask function as opposed to Agent, it seems to still be functioning in an agentic loop. It takes longer to have basic conversations about high level ideas and really kills my experience.

I hope void doesn’t become an agent dumping ground where this behavior is thrust upon the user as much as possible

Not to say I dislike agent mode, but I like to choose when I use it.

When you say "vibe code" do you mean the true definition of that term, which is to blindly accept any code generated by the AI, see if it works (maybe agent mode does this) and move on to the next feature? Or do you mean prompt driven development, where although you are basically writing none of the code, you are still reading every line and maintain high involvement in the code base?

https://open.substack.com/pub/orangepuff/p/first-impressions... I used Claude code to get started on a pdf reader I wanted to build. This pdf reader has a built in LLM chat and when you ask a question about the pdf you’re reading, the page text will be automatically prepended to the question.

Nothing fancy or special. It was built with streamlit in about 150 lines and a single file. But I was impressed that Claude code 1 shot it

May I ask how you got the opportunity to invest in this company? If you are a VC, makes sense, just wondering how normies can get access to invest in companies they believe in. Thanks

I have started using AI for all of my side projects, and am now building stuff almost everyday. I did this as a way to ease some of my anxiety related to AI progress and how fast it is moving. It has actually had the opposite effect; it's more amazing than I thought. I think the difficulty in reasoning about 2) is that given what interesting and difficult problems it can already solve, it's hard to reason about where it will be in 3-5 years.

But, I am also having more fun building things than perhaps the earliest days of my first code written, which is just over 7 years now. Insofar as 1) goes, yes, I never want to go back. I can learn faster and more deeply than I ever could. It's really exciting!

When making requests, does your tool use the normal chrome user agent header or does it specify the request is coming from meha?

I am enjoying Trae so far, but was wondering if it is possible to point it at a local model. https://docs.trae.ai/docs/open-source-software-notice?_lang=... I am not seeing any mention of this in the docs

This is very interesting, but a couple of things to note; 1. o1 still achieves > 40% on the varied Putnam problems, which is still a feat most math students would not achieve. 2. o3 solved 25% of the Epoch AI dataset. - There was an interesting post which calls into question how difficult some of those problems actually are, but it still seems very impressive.

I think a fair conclusion here is reasoning models are still really good at solving very difficult math and competitive programming problems, but just better at ones they have seen before.

I really enjoyed the article, as I am also a nerd who has played runescape for what seems like forever now, and now most of my interaction with the game is via programming. I have two types of bots I use mostly, color bots, where a screenshot is taken and objects are detected by their surrounding pixel colors, and basic click bots. I've actually found that ~100 lines of python code using pyautogui is more than enough to automate tons of annoying aspects of the game.

I am curious, is your Java client one of the many open source bot clients that actually calls into client code? Or is it some type of click script which does some repetitive inputs?

I have had bad luck with the former in terms of getting banned.

As everyone else has already pointed out, better testing would have been very useful here. For instance, print(len(our_ids)) would have been a dead giveaway that that something was up

I am also a junior dev and completely empathize with being given a lot of responsibility and potentially messing up. I think for someone with < 1 year of experience, to solve the problems you created as fast as you did is really impressive. Thankfully your story ends well :)

I don't like ads either, but I also couldn't possibly expect content on the internet to be free, so am happy with the exchange we have currently.

Hello. I just recently built a website blocker Firefox extension so I wanted to check your productivity tool out. The UI is really good and I left it a 5 star.

If interesting in feedback; 1. It seems to only run the block/reroute to to-do list logic when a page is loaded. If the user is already on their site and then adds that site to the blocked list and has to-do list items, it won't be blocked immediately. I don't think this feature is vital but might align better with expectation. 2. Maybe show the to-do list and blocked sites button on the main extension popup, rather than having to go -> to-do list -> blocked sites 3. Maybe on/off button in case user has some sort of issue and needs to quickly access a site that is blocked.

This is my first year participating. I've solved about 20-30 questions from previous years and really loved them. I tried to compete last night for top 100 and was 7 seconds off (rank 123). I shouldn't have tested my code for the first problem because it was too trivial!!

But, that gets the competitive aspect of this challenge out of the way immediately so I can simply have fun with these problems :)

I am excited for them to become a bit more challenging because my friend and I are going to work on them together.

I've had a karat interview before on behalf of Palantir and by the end of the interview I was less interested in Palantir (although thankfully I did move forward) than I was in Karat. I didn't end up looking too much into it. Do you think engineers with ~1 experience could get hired there for interviews? I am about to get promoted to SWE2 if that helps. I actually enjoy leetcode type problems and have solved over 800, so I think this would align with my interest.

I wonder if a social media site marketed as a "free speech" platform will become something center left and right people actually use rather than a haven for psychotic racists. I do not think Gab should be shut down or anything, but just take a look at the average post there.

Maybe all good things simply need to come to an end. I feel like he basically has boomer politics now and the same predictable opinion over and over can get old. My brother said something that I think is true; Joe seems like the kind of guy who believes whatever the last person he spoke to said, so after moving to Austin he's become more conservative.

What we know from this well written and helpful article; The false positive rate Apple told us their algorithm had seems to be accurate. If a machine learning model is extracted from the OS it exists on, it will be much easier to generate adversarial attacks.

For example, a neural net's cost function is just a multivariate function with weights as its input. To figure out how to move those weights (positively or negatively), the gradient of the function is calculated and the weights are nudged in the opposite direction (gradient is the direction of the largest growth of a function, we are trying to minimize the cost). Now, assume we are given a cost function and the weights are constant, now, the input can be the image. So, we take the gradient of the cost function with respect to the image pixels and can now see how we should nudge those to maximize the cost. Apple will absolutely need to protect against adversarial attacks for this to be viable. I'm hopeful.

Today HackerNews learns about adversarial machine learning, which every ML model is susceptible to.

To be clear, that would essentially be working as intended. Apple admitted there would be false positives.

So does this mean that no measure to prevent spread of CSAM should be accepted so long as it has false positives? If we are to trust Apple's numbers, that would be every 1/1trillion images.

Hypothetical; Suppose that this scanning program only ever did what Apple said it was going to do, look for known CSAM. Would this still be upsetting? I am trying to parse if the blow back to this announcement is rooted in the tech communities' ideal of near perfect privacy, or if instead it's a reaction to what this tech could potentially be used for.

I don't find the following argument compelling; Because this tech will be used to scan known CSAM, it will necessarily one day be used to scan for non CSAM. If Apple can implant this tech on your IPhone now, it always could have, and therefore the threat of the government coercing Apple to scan all images for whatever pernicious reasons they can think of has always existed.

CSAM is a massive problem. The solution to how we deal with it will be nuanced and plagued with tradeoffs, but I refuse to be an extremist for either side. I do want something done about CSAM, which is why I am happy that Facebook reports over 10 million instances of it per year from messenger. I also want devices to be mostly private (to assume that a device manufactured by a large corporation would ever be perfectly private in the internet age is delusional). But anyone who acknowledges that CSAM is a problem must also acknowledge that some sacrifice of privacy would be necessary to mitigate it. Or, perhaps one day we can rely on homomorphic encryption to deal with this.

Let C be the point where they meet in the middle and s be dawn.

It takes Lady 1 `12 - s` time to walk AC and 4 time to walk BC. It takes Lady 2 `12 - s` time to walk BC and 9 time to walk AC.

Let's find the ratio of time it takes Lady1 to walk a distance to time it takes Lady2 to walk a distance. The ratio should be the same for all distances since their speed doesn't change.

Thus, (12-s)/9 = 4/(12-s) <=> s^2 -24s + 108 = 0. The quadratic formula gives us s = 6, 18. Since 0 <= s < 12, s = 6.

I wonder if #27 is supposed to be proven without Fermat's Little Theorem. (The question is if p is an odd prime, then 2^{p-1} = pk + 1 for some integer k). Since p does not divide k, it follows from Fermat's Little Theorem that p | (2^{p-1} - 1).

I’m not a denier. I believe AI will vastly alter the way we write code. But, to be honest, it’s very depressing to me. I think I am someone who selfishly enjoys writing code, not necessarily getting software built. If my job became designing something at a very fine grained level, feeding it to an AI, having the AI write it and then code reviewing the AI, I’d just switch careers. Unfortunately for me, I’m super early in my career so I hope I can make enough money in the next 20 or so years such that I can retire young.

Out of curiosity, what is the typical way of building stuff like this? (will dig into the code but might not understand if very well) Do you have to read the youtube html code to figure out which things to hide, or is there a completely different approach?