Show HN: Chorus, a Mac app that lets you chat a bunch of AIs at once https://ift.tt/DV0lakn

Show HN: Chorus, a Mac app that lets you chat a bunch of AIs at once There's so many cool models to try but they're all in different places. In Chorus you can chat with a bunch of models all at once, and add your own system prompts. Like 4o with a CBT overview, or a succinct Claude. Excited to hear your thoughts! https://melty.sh/chorus December 30, 2024 at 03:17AM

Show HN: Beautiful Failed Blackhole Simulation https://ift.tt/N941ckS

Show HN: Beautiful Failed Blackhole Simulation Epilepsy/Seizure Warning: If you have epilepsy or experience seizures, the animation is a failed attempt at recreating the black hole visualization from the film "Interstellar" with three.js that resulted in a plane shaped event horizon phasing through a spherical black body around which point particles revolve. Please skip it if you have a condition that could cause a seizure as a result of looking at any sort of visualization. For everyone that wants to safely check it out, simply do not interact with the screen for a few seconds to make the UI fade away so you can look at the visualization. Cheers everyone. https://mobleysoft.com December 30, 2024 at 02:00AM

Show HN: Handwritten Christmas Card for Hacker News https://ift.tt/GnhLpor

Show HN: Handwritten Christmas Card for Hacker News Hi HN, I’ve been working on a small project that transforms handwritten notes into animated, shareable cards. While the create functionality isn’t live yet, I wanted to share a sneak peek by creating a handwritten Christmas card specifically for the HN community. I started thinking about this after seeing too many AI-generated cards, cookie-cutter email templates, and overly polished designs that lack any personal touch. A friend recently sent me a handwritten card in the mail, and I found it nice that he took his time to write a handwritten note. I wanted to capture that same feeling without the overhead of snail mail. https://ift.tt/HpyB2YT December 30, 2024 at 12:20AM

Show HN: A search engine that values depth over popularity https://ift.tt/Ns1WpoY

Show HN: A search engine that values depth over popularity Hey HN, I'm excited to introduce Graphthem, a search engine designed to explore the deeper layers of knowledge rather than just surface-level popularity. While many AI search engines simply summarize top N results, we've found this approach often misses the many good stuff that is buried deeper in the links and references. Graphthem takes a different approach. we don't just look at the first few pages we find. We also dig into what those pages link to, so you get the whole story. This allows us to deliver answers that capture not just what's immediately visible, but also the foundational ideas and deeper insights that inform those results. Here's a comparison between Perplexity and Graphthem: Query 1: "most watched youtube videos?" Perplexity: https://ift.tt/7lKEDaf... Graphthem: https://ift.tt/djZKqNU... - Query 2: "what is pagerank?" Perplexity: https://ift.tt/wzagYs5... Graphthem: https://ift.tt/OR0FYQg... --- Some related ideas I've explored before: https://ift.tt/1i83Wfx https://ift.tt/0cHV3FE I would love to hear your feedback. Please let us know how we can improve. https://graphthem.com December 29, 2024 at 10:09PM

Show HN: Emacs.tv – A video index of Emacs-related content https://ift.tt/hjldPb6

Show HN: Emacs.tv – A video index of Emacs-related content https://emacs.tv December 29, 2024 at 01:33AM

Show HN: Sine Wave Speech, a real-time audio effect in Rust->WASM https://ift.tt/4zuQpcK

Show HN: Sine Wave Speech, a real-time audio effect in Rust->WASM Source code: https://ift.tt/enT59Rc An interactive website for exploring how human speech can be reduced to just a few sine waves and still be intelligible - or at least recognized to be a human voice. On top of that, I've added various ways to make the sound more musical. The sound effect is based on research from the 80s, which has largely been forgotten by then. To have the effect run in real-time in the browser, I opted to use Rust compiled into WebAssembly. Still, I had to rewrite some linear algebra routines. More about that experience in the blog post: https://ift.tt/1CRE9X7 https://ift.tt/HbOw6lr December 28, 2024 at 09:04PM

Show HN: BashForm – Create and fill forms in your terminal via SSH https://ift.tt/T4vpkqf

Show HN: BashForm – Create and fill forms in your terminal via SSH Instead of boring web forms, create and share forms right in your terminal using SSH. No browsers, no apps, just pure terminal goodness. Key features: - Instant form sharing using simple codes (ssh -t bashform.me f yourcode) - Authentication via SSH keys - Create forms with rich inputs (text, textarea, select) - View responses in the terminal - Zero installation for users - if you have SSH, you're ready - Beautiful TUI (this is subjective xD) Try it instantly: ssh -t bashform.me f try Built with Go and Charm libraries for a snappy, modern terminal experience. P.S. BashForm is actively being developed with new features added regularly. Star the repo to follow the journey and feel free to suggest features! Repo: https://ift.tt/JK85fya https://ift.tt/JK85fya December 28, 2024 at 01:22AM

Show HN: Houseplant – Database Migrations for ClickHouse https://ift.tt/uaVnhfl

Show HN: Houseplant – Database Migrations for ClickHouse https://ift.tt/vC5kh2g December 27, 2024 at 11:41PM

Show HN: Turn Your Study Materials into Interactive Quizzes with AI https://ift.tt/OepKjcu

Show HN: Turn Your Study Materials into Interactive Quizzes with AI Hi Hacker News, I'm excited to share the launch of SyncStudy, a web application that helps students, tutors, and professionals upload study materials, generate quizzes from them, and collaborate with others by sharing these quizzes. It's designed to make learning smarter, more interactive, and more efficient. Key Features: - Upload Study Material: Upload any study document (e.g., PDF, text) and generate quizzes instantly. - AI-Powered Question Generation: SyncStudy uses AI to automatically create relevant questions based on the uploaded content. - Share & Collaborate: Share quizzes with friends, classmates, or colleagues for collaborative learning. - Cross-Industry Use: Although designed for students, the tool is also useful for tutors, trainers, and professionals. Why It’s Different: SyncStudy is not just another quiz generator. It's designed for efficiency—save time by converting study materials directly into valuable quizzes for faster review and learning. Our AI engine ensures that the questions are relevant and cover the material in a meaningful way, which can be extremely useful for exam prep, training sessions, or even creating personalized learning content. We’re currently working on expanding features like: - Support for more file types (e.g., Images) - Integration with other learning tools and platforms - AI-driven suggestions for study improvement If you're interested in learning more, please check out the website: https://syncstudy.app Feedback Wanted! As we're a small team (actually just me at the moment ), any feedback, ideas, or suggestions would be greatly appreciated. I’d love to hear how this could be improved to better serve students, teachers, or professionals in various fields. https://syncstudy.app December 27, 2024 at 05:01AM

Show HN: I made a web app that generates resumes and cover letters for free https://ift.tt/mPlyC41

Show HN: I made a web app that generates resumes and cover letters for free I wanted to share a project I've been working on since October '24. It's an AI-powered toolkit designed to help job seekers shorten their job hunt and advance their careers. The idea came from my own experiences and frustrations with the job application process, both as a job seeker and a hiring manager. I wanted to simplify and streamline how people present their skills and experiences to potential employers. By leveraging AI (GPT-4o), CareerCrate.io can generate tailored application materials in minutes, for free - no sign up required. I'm also developing some optional advanced features like Pursuit™, a tool for job seekers, and SkillForge™, a personalized skill development platform, which will be available with a Pro subscription ($9/month). I'm sharing this with the HN community to get your thoughts and feedback. I'm especially interested in how the AI performs in generating useful content and any suggestions on improving the tool. Please let me know what you think! https://ift.tt/RvwohdZ December 27, 2024 at 04:12AM

Show HN: Quixotic – a tool for wasting bot and LLM scraper time https://ift.tt/1MlXd8E

Show HN: Quixotic – a tool for wasting bot and LLM scraper time I get a lot of bot traffic, most of which doesn't appear to respect robots.txt, so I made a tool to easily rewrite my content to serve to these bots instead. It consists of two components: quixotic - a command line tool that is static-site friendly to generate a copy of a website with some of the words replaced using a Markov generator. linkmaze - a web server that I can send the worst bots to. It generates 100% Markov content on the fly and with random links that also refer to linkmaze content. https://ift.tt/1R64qFb December 26, 2024 at 11:50PM

Show HN: I've made a Monte-Carlo raytracer for glTF scenes in WebGPU https://ift.tt/7jZ5Ibt

Show HN: I've made a Monte-Carlo raytracer for glTF scenes in WebGPU This is a GPU "software" raytracer (i.e. using manual ray-scene intersections and not RTX) written using the WebGPU API that renders glTF scenes. It supports many materials, textures, material & normal mapping, and heavily relies on multiple importance sampling to speed up convergence. https://ift.tt/tJC4SGx December 26, 2024 at 10:54PM

Show HN: Map of YC Startups https://ift.tt/e79L05C

Show HN: Map of YC Startups Hey Everybody! Hope you had a merry christmas Today I had a bit of fun with Claude. Started by scraping YC's startups list, then ran them through OpenAI's embedding service, then UMAP'd the embedding to reduce the dimension to just two coordinates and then just forced Claude to write React that would compile to visualize that. I had fun and I think it's interesting, so take a look! Also note that you won't be able to zoom on mobile (found about this Plotly limitation way too late). If there's interest I can fix this issue by changing plotting libs tomorrow :) Merry christmas https://ift.tt/jn7fCbh December 26, 2024 at 04:07AM

Show HN: Podcast API https://ift.tt/fZgUN6W

Show HN: Podcast API https://ift.tt/3uIEGV1 December 26, 2024 at 12:05AM

Show HN: Homepagr – Bookmarks for Work https://ift.tt/hoyZpQf

Show HN: Homepagr – Bookmarks for Work https://ift.tt/FZ3Bj4C December 26, 2024 at 01:15AM

Show HN: SMS Reminders for Birthdays https://ift.tt/riypFNh

Show HN: SMS Reminders for Birthdays I built a simple service that sends SMS reminders for birthdays. Why? I kept missing birthdays, even with calendar apps. This single-purpose tool is simple and (so far) reliable. Since my texting inbox has a far better signal-to-noise ratio than email or calendar events, it ensures I never miss an important date. Facebook used to serve the same purpose, but it's a dying social graph for me. https://ift.tt/n70lGeU December 25, 2024 at 11:52PM

Show HN: I Ported GHC Haskell Compiler to Windows 11 ARM. MC Gift https://ift.tt/EpFdPcU

Show HN: I Ported GHC Haskell Compiler to Windows 11 ARM. MC Gift Merry Christmas, everyone! Now you can compile Haskell code on Windows 11 ARM. It will run full speed if you use UTM/QEmu on Apple silicon. :-) It's a very draft version, but it works well. Any ideas? https://ift.tt/hOnvyTK December 25, 2024 at 03:11AM

Show HN: Minimalist, text only search engine with prefixes to play with https://ift.tt/Q1YNtX6

Show HN: Minimalist, text only search engine with prefixes to play with I made this classic looking text only search engine(sorry if that's overexaggerating) where you can get concise and straightforward results with sources. There are 5 prefixes right now. It is a prototype kinda thing right now. I am busy for my upcoming exams. So, I would appreciate if anyone is interested to contribute. I am currently using Gemini for this project. If anyone can implement better api's and improve code and the features it would be great help. It's generating fake source links right now, the search is slow, prefixes like weather and time needs to be fixed using api's related to weather and time cause gemini can't give the current time and weather of any place. Pdf resources its giving are fake and do not exist. Let me know what more I can add here. Or directly contribute by yourself. https://ift.tt/t6MlpGb December 25, 2024 at 12:56AM

Show HN: Dynamic RSS Feed Generator https://ift.tt/ycp6KBq

Show HN: Dynamic RSS Feed Generator https://ift.tt/d8Xunjc December 24, 2024 at 11:51PM

Show HN: Otto-m8 – A low code AI/ML API deployment Platform https://ift.tt/JhLXICQ

Show HN: Otto-m8 – A low code AI/ML API deployment Platform Hi all, so I've been working on this low to no code platform that allows you to spin up deep learning workloads(I'm talking LLM's, Huggingface models, etc), interconnect a bunch of them, and deploy them as API's. The idea essentially came up early in September, when experimenting with combining a Huggingface based BERT model with an LLM at work, and I realized it would be cool if I could do that instantly(especially since it was a prototype). At the time, I was considering a platform that could essentially help you train deep learning models without any code. It was my observation that much of the code required to train or even run inference on HF models have matured significantly. But before I solved that problem, I wanted to solve inference. Initially inspired by n8n and AWS Cloudformation, I built out otto-m8 (translates to automate). Given a json payload that lists out all the resources, and how each model is interconnected, launch it as one-off API the user can query. And thanks to Reactflow, the UI was just something I couldn't just not implement. And as I built it out, I did not want to miss out on the LLM and Agent bit. With otto-m8, today, you can launch complex workflows by interconnecting HF models and LLM's(currently it supports OpenAI and Ollama only). But I like to see it being more than that. At the core, every workflow is an input process output model. Inputs get processed and there's an output. Therefore, with the way things are setup, one can integrate almost anything and make it interconnectable. Project Link: https://ift.tt/ITH2QiE Let me know what you guys think. I really would love feedback! https://ift.tt/ITH2QiE December 22, 2024 at 03:09AM