HNHacker News
TopNewBestAskShowJobs

bGl2YW5j

139 karma · joined July 11, 2018

submissionscomments
bGl2YW5j··on Being too ambitious is a clever form of self-sabotage
To you maybe. People get satisfaction and purpose from different things. Unbounded curiosity can often drive tangible outcomes too. You might even have that curiosity to thank for methods and tools you use in your own persuits!
bGl2YW5j··on Writing Code Was Never the Bottleneck
I can’t get over the idea that I won’t ever trust my data to a product made entirely by AI; one with no or limited human oversight.
bGl2YW5j··on What to build instead of AI agents
There's an implication in the messaging of most of these blogs that LLMs and the approach the blog describes, is verging on AGI.
bGl2YW5j··on The new skill in AI is not prompting, it's context engineering
It undermines the credibility of ideas that probably have more merit than this ridiculous labelling makes it seem!
bGl2YW5j··on The new skill in AI is not prompting, it's context engineering
What you're describing sounds a lot like traditional training of an ML model combined with descriptive+prescriptive analytics. What value do LLMs bring to this use case?
bGl2YW5j··on The new skill in AI is not prompting, it's context engineering
Good points that you and Aleksiy have made. Thanks for enhancing my perspective!
bGl2YW5j··on The new skill in AI is not prompting, it's context engineering
Saw this the other day and it made me think that too much effort and credence is being given to this idea of crafting the perfect environment for LLMs to thrive in. Which to me, is contrary to how powerful AI systems should function. We shouldn’t need to hold its hand so much.

Obviously we’ve got to tame the version of LLMs we’ve got now, and this kind of thinking is a step in the right direction. What I take issue with is the way this thinking is couched as a revolutionary silver bullet.

bGl2YW5j··on Gartner Predicts over 40% of Agentic AI Projects Will Be Canceled by End of 2027
If you define success simply as “not cancelled”, then yes, fantastic odds
bGl2YW5j··on Lossless LLM 3x Throughput Increase by LMCache
I’m with you. It’s a bit shocking.
bGl2YW5j··on Lossless LLM 3x Throughput Increase by LMCache
Same. Maintain skepticism.
bGl2YW5j··on IDF officers ordered to fire at unarmed crowds near Gaza food distribution sites
Wikipedia on this topic is incredibly biased to one side or the other. It’s not a valid source in this case.
bGl2YW5j··on Bogong moths use a stellar compass for long-distance navigation at night
The traps themselves. Look up “Brewarrina fish traps”
bGl2YW5j··on TradeExpert, a trading framework that employs Mixture of Expert LLMs
I feel like applying MoE to LLMs is a just a brute-force prediction approach that's neither efficient or very accurate.
bGl2YW5j··on AI helps unravel a cause of Alzheimer’s and identify a therapeutic candidate
Thanks for highlighting this
bGl2YW5j··on Tracing the thoughts of a large language model
Yes. Simply, and well put.
bGl2YW5j··on I genuinely don't understand why some people are still bullish about LLMs
Love the pragmatic and varied use. Nice one and thanks for some ideas.
bGl2YW5j··on I genuinely don't understand why some people are still bullish about LLMs
Can anyone offer insight and/thoughts on his mention of knowledge graphs helping (beyond KGs provide robust source attribution)?

I've been looking at these recently but not specifically to solve "LLM issues".

bGl2YW5j··on Windows Kills SMB Speeds When Using Tailscale
Thanks to the author for this!

What oddly coincidental timing ... I finished setup of Tailscale just yesterday and ran into this exact issue when testing it. I didn't think too much of it and blamed the USB connection I'm using to connect my external drive.

bGl2YW5j··on Scientists re-emerge after a year in Mars simulation project
This is quite a sacrifice for humanity's sake, so thank you to these people for taking part in this experiment.

I'm surprised more effort wasn't put in to making the habitat homely. It's very austere and scientific. I'd have thought that this experiment would also provide means to test ways to provide comfort in trying situations.

bGl2YW5j··on Officially Qualified – Ferrocene
It's not embarrassing at all! I wouldn't have realised had the person not pointed it out. At least you've learnt something interesting.
bGl2YW5j··on Write more "useless" software
I've benefitted from your app! Thank you for the awesome anecdote.
bGl2YW5j··on How to communicate when trust is low without digging yourself into a deeper hole
I find the parent comment insightful, because my immediate reaction to seeing this article headline was, "that's what I needed when dealing with that difficult person recently...".
bGl2YW5j··on Show HN: ListLang – Learn Languages to Fluency
Just downloaded the app... so far it seems like the language learning tool I've been wanting for years. This most-common-vocab-first approach is exactly how I've been learning languages, but have had to make do with the decks available on Anki.

I love that ListLang includes listening exercises too. Vocab + listening is the way to go!

Thank you for this. It's an incredibly polished app so far, very excited to see how it progresses. Ideally you'll avoid Duolingo's heinous use of forced ads. Anki on IOS costs $40 or so. I'd pay that for ListLang provided it proves itself overtime.

What would be helpful: 1. An option to use a classic flashcard format, with one word on it. Fill-in-the-blank can feel a bit hand-holdy. 2. Ability to do listening practice only on sentences learnt so far. I want to practice listening as I learn new words. 3. Sentences in listening practice that use learnt words, but new sentences different to the one used in 'Learn'.

bGl2YW5j··on New neural network architecture inspired by neural system of a worm
I'm interested; could you expand on this please? How can we give electrons more 'freedom' and what would that result in?
bGl2YW5j··on Show HN: Whole Git repo was made with ChatGPT
I spent all weekend playing with ChatGPT.

What I found most powerful was the ability to interrogate its responses. I can ask broad/general questions, and then follow-up with 'why is X used in this way? How else can I do that?' or 'you mentioned Y, how does it work and why isn't X used instead?'.

I had ChatGPT explain to me how Rust's implementation of `Vec` works, and importantly, WHY things have been done in a certain way, e.g. why a separate struct called `RawVec` is used.

This may be simple information, but to be able to get it by asking questions in ways that make sense to ME, is so valuable.

bGl2YW5j··on GitHub Copilot isn't worth the risk
"...does not gives you any social signals lower experienced engineers can use to discern quality or correctness" is very astute.

I experienced this in practice. I was pairing with an inexperienced engineer who was using Copilot. He was blindly accepting every Copilot suggestion that came up.

When I expressed doubt in the generated code (incorrect logic + unnecessarily complex syntax), he didn't believe me and instead trusted that the AI was right.

bGl2YW5j··on A 17-year-old designed a novel synchronous reluctance motor
Insightful comments like this are why I love HN. Thank you!
bGl2YW5j··on Narcissism is rampant, so how do we address it?
Thank you! That does help. I’ll definitely keep these strategies in mind.
bGl2YW5j··on Narcissism is rampant, so how do we address it?
Both this article and your comment sum-up what I have really been struggling with lately at work. I’ve never experienced such a narcissistic person and have had a hard time not only dealing with their behaviour, but identifying it for what it is in the first place.

Thanks for the insight.

bGl2YW5j··on Ask HN: How do you deal with rude interviewers?
The majority of references I could find were on Urban Dictionary and other ‘how to not be a beta’ type sites. Hardly reputable sources or places that stand up for women.
← PreviousPage 2 of 3Next →