crackalamoo

8 ms·

61.
by crackalamoo 2y ago
o1 could be useful for heavy math/science/coding work, such as a study tool or coding assistant. If new o1-based tools come out and the model gets too good at this kind of work, many software engineers might be cooked. The codefor
62.
by crackalamoo 2y ago
Not a solution, but maybe if a bad actor tried to create a bioweapon, a trusted organization could use this technology as an antidote. Unfortunately this still leaves the possibility of some kind of insidious, undetectable bioweapon.
63.
by crackalamoo 2y ago
It comes and goes. I always know most of the syntax of languages I know, but with a bias towards languages I've been using recently or very familiar with. For example, I don't know C++ syntax that well off the top of my head right
64.
by crackalamoo 2y ago
In my mind, it's not about medical evidence for dark mode: I just prefer the experience as a user. Also even light mode is often gray on gray, with backgrounds like #eee and text like #112 for example. The contrast is probably lower on
65.
by crackalamoo 2y ago
I disagree overall. I find it much less painful on the eyes to read in dark mode at night, even if there's no evidence for health benefits. I agree that dark mode as an additional UI shouldn't be an afterthought, and should be tes
66.
by crackalamoo 2y ago
Interesting. But I wonder what simply scaling the model will do if we use the same kind of transformer model and the same kind of data (we might even run out of internet data). Of course, we will probably see non-scaling innovation at least
67.
by crackalamoo 2y ago
Urdu uses a QWERTY-based layout (so ق is where Q normally is, etc.), and adapts the shift key for additional characters. I find this much easier than the Arabic keyboard layout. I think non-Latin script alphabets should try to match QWERTY
68.
by crackalamoo 2y ago
Such a fascinating language! Great example of how linguistic diversity at the extremes can help us understand the human mind better.
69.
by crackalamoo 2y ago
Seconding DigitalOcean, it's worked well for me.
70.
Why scientists are trying to reengineer the cow's stomach(washingtonpost.com)
4 points by crackalamoo 2y ago|1 comments
71.
by crackalamoo 2y ago
Scientists in California are working to genetically engineer the cow microbiome -- and in the process, eliminate methane emissions.
72.
by crackalamoo 2y ago
Really interesting idea overall. But I disagree with this part: > Algorithms are tiny parts of large systems. The algorithm part can usually be abstracted as an API call that provides some value(s) with inherent uncertainty; an ML endpoi
73.
by crackalamoo 2y ago
Maybe this is why it's so hard to find a white collar job. Companies are realizing they don't need an expert in meetings, so they cut staff without hiring creative individuals in their place to do something better than meetings.
74.
by crackalamoo 2y ago
I don't know about corporate hackathons, but I think college hackathons are more than a distraction. I've learned useful technical skills from hackathons (including React). Most importantly, I've been able to reference my exp
75.
by crackalamoo 2y ago
In principle, yes. In practice, it's a little more tricky. AI APIs give mostly human-level reliable results in many cases. But they don't work well 100% of the time, which affects trust in an automated B2B product meant to replace
76.
by crackalamoo 2y ago
How many of these are due to LLM wrappers without a clear advantage over ChatGPT? This AI bubble is like the dot com bubble, where tons of "AI-first" companies will come up and go down quickly. The companies that create genuine ma
77.
by crackalamoo 2y ago
Nice article. I have some questions: Do you think an AI scientist would be a set of decentralized models, or a big foundation model that only a few companies can make, like we see with LLMs? To what extent will an AI science replace human s
78.
by crackalamoo 2y ago
This reminds me of the Microsoft PowerPoint design ideas feature, which is just awesome. Takes the content and puts it into a great design very intelligently. It might just be my habit, but I prefer designing directly in HTML/CSS than
79.
The Future of Humanity's Energy No One Knows About – Terraform(youtube.com)
7 points by crackalamoo 2y ago|0 comments
80.
by crackalamoo 2y ago
Yes, I have seen many of the interviews from these extreme outliers. But I am curious to hear regular people's experiences with luck, and whether they have succeeded without much of it, even if they didn't become billionaires.
81.
by crackalamoo 2y ago
In what way is luck important? One part of me says someone that identifies an urgent market need and executes on it doesn't need too much luck. On the other hand, maybe luck is important in terms of your network: meeting the right pers
82.
by crackalamoo 2y ago
Let's say, hypothetically, we do enough RLHF that a model can imitate humans at the highest level. Like, the level of professional researchers on average . Then we do more RLHF. Maybe, by chance, the model produces an output that is a
83.
by crackalamoo 2y ago
Relevant xkcd: https://xkcd.com/1758/ Dark matter really is a theory with falsifiable data. I'm not sure of the exact arguments, but I believe the evidence available (such as the evolution of galaxies) is only con
84.
by crackalamoo 2y ago
I just use Vim. No reason to get Neovim when it's effectively the same for my usecase, just with extra steps. Vim is already installed anywhere and feels super light. If I wanted a fancy editor, I would just use VSCode (which I do). Vi
85.
by crackalamoo 2y ago
This makes sense to me. I think having components trained in-house is pretty important to building a moat, and probably requires a somewhat successful business beforehand to collect proprietary data. I'm a little doubtful of the value
86.
by crackalamoo 2y ago
This is true, but I also wonder if the fact that AI is new means that there is a lot of low-hanging fruit with applications of AI to new niches. I wonder what the best way is to identify and contribute to these niches, given that an AI solu
87.
by crackalamoo 2y ago
It seems that YC is funding a lot of these companies despite all that. I've been wondering why, because I agree with you about the current capabilities of AI. Is it * they expect AI to drastically improve, making these GPT-based produc
88.
by crackalamoo 2y ago
I'm curious, why does having a PhD help you be open minded in a young and naive way and immerse yourself in the customer space? Or is it that fresh PhDs without much experience are still young and naive?
89.
Ask HN: The Problem with "AI Startups"?
17 points by crackalamoo 2y ago|28 comments
90.
by crackalamoo 3y ago
Is this mainly overuse of trendy packages and frameworks through npm and similar? What exactly is bad about writing JavaScript as an employee?

More ›