15:[[44712758,{"by":"taikon","descendants":2,"id":44712758,"kids":[44712832],"score":3,"time":1753721861,"title":"How to Setup a Secure Home Server: A Complete Guide","type":"story","url":"https://www.davidma.co/blog/2025-07-24-home-server/"}],[44269756,{"by":"7402","descendants":196,"id":44269756,"kids":[44272367,44270303,44271341,44273604,44276554,44272754,44273053,44271270,44284397,44273801,44273805,44275544,44274066,44273436,44273702,44273986,44280402,44271062,44273141,44271703,44271084,44270070,44280999,44271215,44326746,44272662,44277555,44270195,44279205,44272850,44299153,44280159,44272315,44272598,44298232,44279266,44271523,44277560,44299049,44271840,44288921,44271162,44272854,44272893,44270729,44287779,44279177,44289753,44270970,44269918,44273094,44271473,44316933,44272982,44273868,44270262,44322285,44295257,44269956,44275311,44275582,44273634,44271664,44272607,44275786,44273247,44271288,44272787,44271251,44271472,44285234,44284371,44269833,44274251,44271567,44272483,44273685,44279741,44272423,44271273,44284585,44271337,44272875,44272791,44284127,44296893,44271574,44295780,44271033,44273011,44271349],"score":95,"text":"For the sake of argument, assume that if your job consists of sitting at a computer, reading on a screen, and typing on a keyboard, then your career will go away.

There is always room at the top, and there may always be room for humans at the top of any career. Assume (this is a tough ask, I know) that you are NOT one of those people.

What is your fallback job? What skills do you have or would like to acquire that might keep you going? Bicycle mechanic? Teach music to children? Woodworking/carpentry? (Living off your stock options or investments does not count)","time":1749830604,"title":"Ask HN: What is your fallback job if AI takes away your career?","type":"story"}],[44202103,{"by":"philipisik","descendants":3,"id":44202103,"kids":[44202401,44217396],"score":15,"text":"$16","time":1749225230,"title":"Tiptap open-sources 10 formerly Pro extensions under MIT license","type":"story"}],[44105412,{"by":"akoenig","descendants":43,"id":44105412,"kids":[44106484,44106297,44108213,44106798,44108922,44107405,44106703,44110698,44106831,44113017,44108072,44109009,44108573,44116216,44108732,44105413,44112745,44120131,44108754],"score":156,"time":1748339404,"title":"LiveStore: State management based on reactive SQLite and built-in sync engine","type":"story","url":"https://livestore.dev"}],[43762409,{"by":"akadeb","descendants":94,"id":43762409,"kids":[43763204,43763477,43763898,43763312,43763640,43763765,43763552,43762410,43769886,43763331,43764043,43764414,43763970,43764815,43779959,43765334,43763202,43773572,43764136,43763500,43763140,43763345],"score":177,"text":"$17","time":1745331059,"title":"Show HN: I open-sourced my AI toy company that runs on ESP32 and OpenAI realtime","type":"story","url":"https://github.com/akdeb/ElatoAI"}],[43656476,{"by":"vectify_AI","descendants":1,"id":43656476,"kids":[43656505],"score":11,"text":"$18","time":1744393747,"title":"Show HN: RAG, No Vectors","type":"story","url":"https://github.com/VectifyAI/PageIndex"}],[43118514,{"by":"themanmaran","descendants":40,"id":43118514,"kids":[43154799,43155742,43153303,43154779,43154387,43133356,43131719,43160802,43156354,43154968,43192538,43157691,43161193,43120771,43160964,43192541],"score":146,"text":"$19","time":1740077369,"title":"Show HN: Benchmarking VLMs vs. Traditional OCR","type":"story","url":"https://getomni.ai/ocr-benchmark"}],[42759588,{"by":"babuskov","descendants":11,"id":42759588,"kids":[42760829,42765287,42768957,42760823,42766438,42764084],"score":12,"text":"$1a","time":1737309852,"title":"OpenAI crawler burning money for nothing","type":"story"}],[42585009,{"by":"anshrathodfr","descendants":293,"id":42585009,"kids":[42592245,42593934,42594566,42593863,42594136,42592899,42592274,42592765,42591600,42593735,42593691,42592094,42592706,42594452,42592083,42591793,42593756,42592915,42591841,42593532,42595541,42591725,42593982,42593533,42593229,42596174,42602680,42594538,42597915,42594454,42604672,42599167,42592049,42593393,42591590,42598180,42593793,42599249,42593784,42593231,42594915,42592236,42592987,42593256,42593865,42593413,42594266,42593505,42592405,42596521,42597398,42593151,42602421,42593740,42593066,42592980,42596523,42594534,42594409,42594711,42593292,42597304,42592626,42599060,42593335,42593466,42593944,42596216,42591964,42594787,42591993,42592432,42593264,42597745,42592092,42592675,42591819,42645579,42618520,42592046,42592962],"score":526,"text":"Hi, I'm a developer and first time i shipped the real product after observing the startups and indie hackers community for years.

I had made so many useless apps [you should check my website https://ansh.life], but this time I built a very useful product that has a number of super easy-to-use tools in one app for video, music, and photo files. Users can compress, convert, resize, and do so much more with easy-to-use tools.

Background: I developed a frame-by-frame video cropper to upload cropped landscape videos to Instagram Reels. However, it required FFmpeg, and as a noob video editor, I decided to incorporate more user-friendly video tools. I then introduced image and audio tools to maximize the capabilities of FFmpeg. I use my app daily, and it has surprisingly generated a few thousand dollars for me.","time":1735906827,"title":"Show HN: I completed shipping my desktop app","type":"story","url":"https://pimosa.app/"}],[42323077,{"by":"jdcampolargo","descendants":7,"id":42323077,"kids":[42323548,42327635,42326468,42329754,42323103],"score":6,"text":"I'm working on a large web scraping project and looking to avoid building custom scrapers for every site using something like BeautifulSoup.

Anyone have recommendations for up-to-date scraping APIs or platforms that work well with a variety of websites?","time":1733353934,"title":"What Are the Latest Scraping APIs or Services/Websites?","type":"story"}],[42089719,{"by":"speckx","descendants":4,"id":42089719,"kids":[42090173,42091629],"score":20,"time":1731093903,"title":"Seamlessly transition your Twitter network to Bluesky with just a few clicks","type":"story","url":"https://skyfollowerbridge.com/"}],[41715554,{"by":"iknownthing","descendants":4,"id":41715554,"kids":[41716090,41715871,41717380],"score":3,"text":"Basically I have a use-case where I am building an llm-based tutor where some questions you ask the tutor can be adequately answered by gpt-4o-mini but others required gpt-4o. I'm looking for a router that can route to gpt-4o only when necessary and otherwise route to the cheaper gpt-4o-mini. First of all, is this even something that can be done reliably at this point? If so, are there any recommendations for LLM routers that can do this?","time":1727825727,"title":"Ask HN: LLM Router Recommendations","type":"story"}],[41249044,{"by":"titusblair","descendants":2,"id":41249044,"kids":[41251457],"score":16,"text":"$1b","time":1723659959,"title":"We Built a Tool to Hack Our Own AI: Lessons Learned Securing Chatbots/Voicebots","type":"story"}],[40621466,{"by":"stevofolife","descendants":5,"id":40621466,"kids":[40631014,40621684,40626488,40626548,40621590,40621553],"score":2,"text":"With the rise of large language models (LLMs), what’s the easiest way to scrape the web today? Additionally, what’s the simplest method to automate interactions with a website?","time":1717898203,"title":"Ask HN: What is the easiest way to scrape websites today?","type":"story"}],[37931688,{"by":"maxkrieger","descendants":2,"id":37931688,"kids":[37942794,37932989],"score":14,"text":"There's a lesser-known trick described in the OpenAI Cookbook. You can linearly transform a generic embedding space (like OpenAI’s) for your usecase with a well-tuned matrix multiply.

I ported the cookbook technique into a standalone hardware-accelerated webapp, and added some UX to guide the process along. There are some examples on the homepage for out-of-the-box fun. With 30 seconds of training, I was able to improve performance by the first example by 15%.","time":1697649359,"title":"Show HN: Finetune OpenAI Embeddings In-Browser","type":"story","url":"https://a9.io/spaceshifter/"}],[37338796,{"by":"shawnxman1423","descendants":2,"id":37338796,"kids":[37338907,37338892],"score":2,"text":"Hey guys, \nRecently I created an internal tool for my team and I that write notes from all of our meetings and save them into a spreadsheet, following the "The Mom Test" book guidelines.

Also it shows you some red flags about stuff that happens during conversations such as fluff, bad questions, etc...

What do you guys think? \nHow do you write and analyze notes today?","time":1693495961,"title":"MomTestGPT – A Better way for founders to write notes","type":"story"}],[36794045,{"by":"tikkun","descendants":3,"id":36794045,"kids":[36794427,36794046],"score":3,"time":1689803849,"title":"Llama 2 Resources","type":"story","url":"https://gpus.llm-utils.org/llama-2-resources/"}],[36566529,{"by":"zoe_dk","descendants":5,"id":36566529,"kids":[36566979,36568895,36566666],"score":8,"text":"Eddie the Smart Video Editor is a Python script that automates the process of editing videos based on specific code-words you say while recording the video.

It takes a video file, converts it into an audio file, transcribes the audio to text using OpenAI's Whisper API and identifies segments between specified code-words. It then cuts out these segments from the original video, removes any silence and produces a new edited video file ready for the final touches.\nFalse recording starts are also okay as it only matches the last occurrence of the 'start' code-word when the next code-word is the 'end' code-word.

I wanted to jump into the far deepest of deep ends and ask for feedback please.

I am kind of scared of you all tbh but I value learning more than crapping my pants.

Thank you.","time":1688337480,"title":"Show HN: Eddie – A Smart Video Editor with Python [Seeking Feedback]","type":"story","url":"https://github.com/ZoeDekraker/eddie-smart-video-editor"}],[36265931,{"by":"duarteoc","descendants":1,"id":36265931,"kids":[36277337],"score":3,"time":1686354668,"title":"Supercharging my Telegram group with the help of ChatGPT","type":"story","url":"https://duarteocarmo.com/blog/supercharging-telegram-bot-chatgpt-python"}],[35864102,{"by":"david-venegas","descendants":7,"id":35864102,"kids":[35864104,35864139],"score":5,"time":1683565508,"title":"I Made a ChatGPT Microphone Chrome Extension (Open Source)","type":"story","url":"https://chrome.google.com/webstore/detail/chatgpt-microphone/kpnejlajlhnpfphbhgfipmlogplmidin"}],[35859281,{"by":"NKosmatos","descendants":47,"id":35859281,"kids":[35861018,35860108,35862554,35860116,35862074,35862157,35860135,35860325,35860324,35862186,35862938],"score":107,"time":1683534077,"title":"DJI drone flight log viewer","type":"story","url":"https://www.flightreader.com/"}],[35852928,{"by":"krishadi","descendants":2,"id":35852928,"kids":[35853202],"score":1,"text":"We've been developing prompts for writing agents with OpenAI. After working on it for the last months, we have a huge amount of prompts lying everywhere. We don't have a good workflow around keeping tracking of prompts, scoring them, revisiting them later on, tacking their output and degradation to give the right output over time.

How do people here do this? What is your workflow around this?","time":1683475944,"title":"Ask HN: How do you evaluate prompts?","type":"story"}],[35344294,{"by":"giuliomagnifico","descendants":54,"id":35344294,"kids":[35344350,35346663,35344499,35350505,35351068,35351081,35346679,35344685,35344362],"score":48,"time":1680023030,"title":"Twitter is dying","type":"story","url":"https://techcrunch.com/2023/03/28/twitter-is-dying/"}],[35085944,{"by":"dpatel1","descendants":2,"id":35085944,"kids":[35085979],"score":1,"text":"We are Wana - a hobbyist’s ultimate tool for learning, creating, and connecting. Our mission is to help others build confidence and self-love through their unique sparks. Naval’s mindset has been a huge motivator for us to keep going as it aligns perfectly with our goals. Any help is appreciated!","time":1678392421,"title":"How can I reach Naval Ravikant?","type":"story"}]]

user:	gbertb
created:	September 19, 2012
karma:	12