Sai_

24 ms·

211.
LLMs open up new paradigms in Human Interface Design (HID)(pretzelbox.cc)
4 points by Sai_ 3y ago|0 comments
212.
by Sai_ 3y ago
Where do you see embeddable chats being valuable? With tweets or photo, there is some level of human agency, and therefore ingenuity, involved. Here it’s just prompt engineering.
213.
by Sai_ 3y ago
I meant the hype around open source LLMs, not OpenAI's LLM. On reading your response and my original comment, I suspect you thought I was including OpenAI's LLM as a hype-driven product. Sorry if you didn't think that. That s
214.
by Sai_ 3y ago
Could you explain what other choices were red flags for you? I’m somewhat familiar with the open source LLMs space but not enough to know why some choices are better than others.
215.
by Sai_ 3y ago
And neither is anyone who has played with these new LLMs, found them so-so, and wondered whether the hype was warranted.
216.
by Sai_ 3y ago
Yes, langchains and llama-index are both amazing techs. They are separate from plug-ins though. Plug-ins are ChatGPT specific (afaik) and are driven by free form content written by plug-in developers for ChatGPT. Langchains start with human
217.
by Sai_ 3y ago
> That is a rather bold statement without proper evidence. I based this on a couple of insights 1. OpenAI itself encourages adding examples to prompts where necessary. Given the same initial prompt, this means that the prompt with exampl
218.
by Sai_ 3y ago
I read your post but I'm not sure how async work relates to Context Warehousing. Maybe this was lost on me because I don't have context (hah) about your writing - maybe you often write about async work and this was a continuation
219.
Understanding ChatGPT prompt engineering through plug-ins manifests(pretzelbox.cc)
1 points by Sai_ 3y ago|4 comments
220.
by Sai_ 3y ago
To me, the plugin model allows two things 1. single point of entry: you don't need ten different apps to achieve ten different outcomes. You can do it all right inside ChatGPT 2. Since I can have up to three plugins active inside ChatG
221.
by Sai_ 3y ago
If free human generated content dries up for OpenAI wouldn't it dry up for Google (both Search and Bard) and Bing as well?
222.
by Sai_ 3y ago
How so? Unless, I'm missing something, the spec you expose to ChatGPT only tells them which api endpoint to hit. The code powering that endpoint is not visible to ChatGPT. The only argument you could make is that you are giving it mach
223.
by Sai_ 3y ago
> Competing app stores are a thing This is only a factor if the underlying AI/LLM model becomes as good as OpenAI AND serves millions of people. While open source LLMs are soon going to gain parity with GPT-4/Bard, it is unlike
224.
by Sai_ 3y ago
I think OpenAI is building a moat right in front of us through their plug-ins idea. Executed well, plug-ins could become ChatGPT's App Store equivalent. Once that happens, OpenAI is undoubtedly going to convert a portion of search traf
225.
Understanding ChatGPT prompt engineering through plug-ins manifests(pretzelbox.cc)
3 points by Sai_ 3y ago|0 comments
226.
by Sai_ 3y ago
This is an interesting question. To paraphrase as per my understanding of your comment, is intelligence an emergent property of being able to interact with each other through language? Say I speak gibberish (to you) which is actually me exp
227.
by Sai_ 3y ago
India really struggles in this regard. We don’t have a widespread pub/bar culture so everyone meets out on the streets…just knots of people sitting on bikes and chatting with each other. This is traffic nuisance but also makes the stre
228.
by Sai_ 3y ago
This sounds suspiciously like a “no true Scotsman” argument to me. Remote first is a spectrum and dismissing lived experiences as “oh, you just haven’t experienced the right kind of remote first” helps no one. What makes you think documenti
229.
by Sai_ 3y ago
I was hired into a senior role (architect level) by a company that was compelled to go remote because of the pandemic. Remotely building working relationships at that level was hard AF since most people I had to work with already knew each
230.
by Sai_ 3y ago
DRI - directly responsible individual, for those who didn’t know.
231.
by Sai_ 3y ago
You must be from a planet with very long years! There is no way I can even begin to digest what you have said in your comment.
232.
by Sai_ 3y ago
Maybe ELI5 was the wrong level for my ask. I still want to understand it from a technical perspective. Maybe ELI am a software engineer but not in AI/ML?
233.
by Sai_ 3y ago
I first heard about him after an internal post/rant of his about Google+ got leaked and went viral. My impression is that he is a Kardashian - famous for being famous - in the software world.
234.
by Sai_ 3y ago
Enlightening example of having a conversation. Makes thing clearer.
235.
Ask HN: Can someone ELI5 transformers and the “Attention is all we need” paper?
644 points by Sai_ 3y ago|228 comments
236.
by Sai_ 3y ago
I think Sindri and Chaunsa are also available in India (Indian here) but I personally think Alphonsos are much richer in taste and texture. Ymmv. Edit - maybe I’m biased because Alphonsos are pricier so the palate thinks it’s getting someth
237.
by Sai_ 3y ago
Having lived in the States and only eaten mexican mangoes for years, returning to India and eating Alphonso and other Indian mangoes quickly showed me what I was missing. Don’t know if that epiphany is worth a 4x price difference but it’s n
238.
by Sai_ 3y ago
Even more than Java Applets, I remember looking at Java Beans and being heavily, mightily underwhelmed.
239.
by Sai_ 3y ago
Clearly your WhatsApp hasn't been inundated with offers from a proliferation of Jio* named companies. They're monetizing WhatsApp for sure - it's just that they are monetizing it through intermediaries offering transactional
240.
Ask HN: What ChatGPT plugins would you like to see?
2 points by Sai_ 3y ago|0 comments

More ›