aquajet
Shellular: Run agents, terminals and browser DevTools from your phone
Re:Anthropic's open-source framework for AI-powered vulnerability disc...
on Thursday June 04, 2026 @07:20PM (#48406000)
Using something like pi helps. I've made my own dotfiles for skills/extensions I like and can install them just like my normal dotfiles
VercelNet: The internet but only Vercel apps
Re:MoltHub-A site where AI agents come to compute (if you know what I ...
on Saturday January 31, 2026 @12:50AM (#46833866)
it's crazy how good AI frontend design and humor has become.
Trusting Driverless Cars (2019)
Percieved Age vs. Actual Age
OpenAI: The Game
Show HN: Wordle but Using Common Mispellings
https://www.sitebrew.ai/BkziZd
I was playing around with AI text -> code generators, and asked it to "make wordle but if a drunk person came up with it". It generated all the code for this, the only extra prompting I needed to do was to refine the word list.
Show HN: I made an accurate YC interview simulator
https://twitter.com/thiteanish/status/1832905181278941610
I created a YC interview simulator based on my past experiences interviewing. The hardest part for me has been tripping on hard questions, and the rapid pace of the interview.So, I focused on making a simulator that helps me with that :)Features: - asks good questions - rapid fire format - records + transcribes your responses
Show HN: Using LLMs to Debate Topics
Hey!This is my weekend project. Given a topic one LLM will argue for it, and the other will argue against it.After 10 turns a judge LLM decides the winner, explains why, and highlights some of the interesting points
Re:A personal AI chess coach
on Friday February 16, 2024 @02:17PM (#39401673)
Cool! How does it work?
Re:Show HN: Using GPT and Reddit for Hyper-Specific Book Suggestions
on Thursday February 15, 2024 @05:23AM (#39381027)
i take some adderall and use the web app a bunch
Show HN: Using GPT and Reddit for Hyper-Specific Book Suggestions
I've been working on a project to use GPT to use Reddit threads + internal GPT knowledge to create book recommendations.GPT operates a browser to search + read reddit threads, extract out relevant recommendations, and show direct links to those books.It also generates follow-up questions to help further narrow down theWhat I found really cool is that it works well for hyper-specific queries...
Re:Google's New AI-Powered Browser Could Mark the End of the Human Int...
on Friday January 26, 2024 @05:01PM (#39149200)
ByteDance is using OpenAI API to build a competitor
Re:OpenAI delays launch of custom GPT store until early 2024
on Friday December 01, 2023 @08:16PM (#38494814)
but the president was the CTO of Stripe
Re:Ultra-processed food linked to higher risk of depression, research ...
on Monday September 25, 2023 @05:54PM (#37651249)
Hope you feel better soon :)
But this shows that theres no eating processed food --> get depressed causation, just a correlation. Otherwise a perfectly happy person who eats a lot of tv dinners would inexplicably get sad with nothing else in their life changing.
Re:Ultra-processed food linked to higher risk of depression, research ...
on Monday September 25, 2023 @05:02PM (#37650556)
From article:
```
Others urged greater caution. Prof David Curtis, an honorary professor at University College London Genetics Institute, said: “The only foodstuffs which [this study] shows are associated with increased risk of depression are artificial sweeteners. Of course, this does not mean that an effect of artificial sweeteners is to increase depression risk – it is just that people with increased risk of developing depression tend to consume larger quantities of artificial sweeteners.”
But the authors disagree. Prof Andrew T Chan, chief of the clinical and translational epidemiology unit at Massachusetts general hospital and co-author of the research, said: “The strength of our study is that we were able to assess diet several years before the onset of depression. This minimises the likelihood that our findings are simply due to individuals with depression being more likely to choose ultra-processed foods.
```
Seems to be a correlation study to me. I also don't see any controls for other circumstances, so it can also be explained that other factors could contribute both to eating processed foods and getting depressed (ex: working long hours, having low income, relationship stress)
Show HN: Use LLMs to increase E-commerce store's conversion rates
Hi everyone! I’m working on Claros, a plugin for e-commerce sites to increase site conversions.Claros asks the customer questions and guides customers to the best product, acting kinda like a physical store associate. This results in more purchases.I use LLMs and hybrid vector search to do the recommendations and the product retrievalLet me know if you would like to try it out for free, or if...
Re:Arwes: Futuristic Sci-Fi UI Web Framework
on Friday June 23, 2023 @04:17PM (#36451709)
examples?
X/Twitter will start paying creators for ads served in their replies
Re:Downtown SF has 18.4M square feet of empty office space. We mapped ...
on Tuesday May 09, 2023 @03:07PM (#35878985)
When I worked in Seaport i'd just walk from south station
Ask HN: How Do You Tinker?
How do you come up with ideas, how do you prioritize what to do, how scrappy are you?I ask because I want to tinker more
Re:Money market funds swell by over $273B as investors pull deposits f...
on Sunday March 26, 2023 @01:41PM (#35317179)
Archive: https://archive.is/JIFwG
Sparks of Artificial General Intelligence: Early Experiments with GPT-4
Show HN: PACman: Automatically convert prompts to Python functions
https://github.com/anishthite/pacman-prompts
I made a package to automatically convert prompts stored in a yaml file to pure python functions.A lot of prompt-heavy code is hard to grasp due to mixed use of large multi-line strings and orchestrating code. Some prompt libraries automate portions of these away, but we lose flexibility in prompting.By extrapolating the prompt config to a yaml file and generating pure Python functions from it,...
Re:LLaMa running at 5 tokens/second on a Pixel 6
on Wednesday March 15, 2023 @02:53PM (#35173018)
I would have tried it but I didn't have enough storage on my phone to hold both the original and quantized weights.
Re:LLaMa running at 5 tokens/second on a Pixel 6
on Wednesday March 15, 2023 @02:35PM (#35172723)
I'm the original tweet author.
Currently typing this from my Pixel after running it countless times :)
Re:OpenAI Is Now Everything It Promised Not to Be: Closed-Source, and ...
on Wednesday March 01, 2023 @02:45PM (#34987296)
How were PCs able to work on a smaller scale? I've been looking into the chips/model trends and I can't see local models that perform at the same level as GPT happening in the near future.