Show HN: I Built a Multimodel AI Chatbot to Avoid Having FOMO on the Best Models Hi HN, I built ChatOne while working on a project and constantly switching between AI models like GPT-4 and newer ones like Claude 3.5. I kept wondering if I was missing out on better answers, so I created ChatOne to get responses from multiple models at once and compare them easily. -Teddy https://chatone.io September 29, 2024 at 09:38PM
Show HN: Modern Benchmarking Tooling for JavaScript I always had a sweet tooth for how easy it is to use google/benchmark, but when working with js, current libraries didn't feel right and some were not even accurate enough, so I decided to create my own library to make JavaScript benchmarking tooling better. With more free time, I finally implemented all features I wished for in 1.0.0 and made a lightweight C++ single-header version for moments when google/benchmark is too much. Hope this library helps you as much as it does me. https://ift.tt/uvjWQBs September 29, 2024 at 12:24AM
Show HN: I Made a Site for Designers to Gen Great Patterns Quickly, No Headaches It was a crisp autumn evening when I found myself burning the midnight oil, yet again. As a graphic designer, I was knee-deep in a project that required intricate patterns for a client's new product line. The clock ticked mercilessly as I struggled with complex design software, trying to bring my vision to life. That's when it hit me – why couldn't I just tell a computer what I wanted and have it create the pattern for me? Why did I need to wrestle with technical jargon and complicated prompts when I had a clear picture in my mind? This revelation sparked a journey that would consume the next eight months of my life. I dove headfirst into the world of AI, determined to create a tool that could understand designers' needs in plain English. It wasn't easy – there were countless late nights, gallons of coffee, and moments of doubt. I experimented with various AI models, eventually settling on a combination of Stable Diffusion and Flux, fine-tuned to understand design language and generate beautiful patterns. The result? AI Pattern – a game-changer for designers everywhere. Here's what makes AI Pattern special: 1. Speak Your Mind: Just describe what you want in everyday language. No need for complex prompts or technical terms. 2. Lightning Fast: Generate stunning patterns in seconds, not hours. 3. Intuitive Interface: User-friendly design that feels like chatting with a creative assistant. 4. Endless Variety: From geometric to organic, abstract to realistic – if you can imagine it, AI Pattern can create it. 5. Customization Options: Tweak and refine generated patterns to match your exact vision. Now, I'm thrilled to share AI Pattern with fellow designers and creative minds. It's more than just a tool – it's a partner in creativity, ready to transform your ideas into visual masterpieces. I invite you to try AI Pattern and experience the joy of effortless pattern creation. Your feedback is invaluable as we continue to refine and improve this design revolution. Together, let's push the boundaries of what's possible in the world of pattern design! So, are you ready to turn your words into wonders? Give AI Pattern a spin and let your creativity soar! https://ift.tt/AukwWZi September 28, 2024 at 05:31PM
Show HN: DecisionBox – Continuous Accuracy Improvement for LLM Apps Just released DecisionBox, an open-source SDK that helps developers make high-accuracy decisions in LLM apps, which continuously improve with more data. DecisionBox tackles the challenge of maintaining decision accuracy beyond the limitations of prompt engineering. It streamlines the data science process with an easy-to-use API and enables ongoing accuracy metrics and model improvements. Get started: https://ift.tt/mY9GlpB We’re excited to hear feedback and answer any questions! AMA. https://ift.tt/mY9GlpB September 28, 2024 at 12:27AM
Show HN: A tool that turns everyday computers into your own AI cloud I have a favour to ask. I’ve been working for a while on Kalavai, a project to make distributed AI easy. There are brilliant tools out there to help AI hobbyists and devs on the software layer (shout out to vLLM and llamacpp amongst many others!) but it’s a jungle out there when it comes to procuring and managing the necessary hardware resources and orchestrating them. This has always led me to compromise on the size of the models I end up using (quantized versions, smaller models) to save cost or to play within the limits of my rig. Today I am happy to share the first public version of our Kalavai client (totally free, forever), a CLI that helps you build an AI cluster from your everyday devices. Our first use case is distributed LLM deployment, and we hope to expand this with the help of the community. Now, the favour! I’d love for people interested in AI at scale (bigger than a single machine) to give it a go and provide honest feedback. Do you share our motivation? If you tried Kalavai, did you find it useful? What would you like it to do for you? What are your painpoints when it comes to using large LLMs? https://ift.tt/YW87g3m September 27, 2024 at 11:42PM
Show HN: 1Minute Break – VS Code extension for developer well-being As a developer, I often found myself working for hours without breaks, leading to eye strain and back pain. I created 1Minute Break to solve this problem for myself and others. that reminds developers to take short, regular breaks. Features include: - Customizable break intervals - Interactive break reminders with suggested activities - Water intake tracking - Ergonomic exercise prompts It's designed to improve productivity and health without disrupting workflow. Feedback welcome! https://ift.tt/hqKTRCE... https://ift.tt/paKChtO September 26, 2024 at 11:29PM
Show HN: FastIndex, an open-source search engine indexation tool There's a lot of paid search engine indexation tools out there and I wanted to create my own. Been working as an engineer for over a decade now and my open-source contribution has always been something I wanted to do. Thus I decided to create FastIndex, an open-source search engine indexation alternative to paid solutions such as TagParrot, URLMonitor, Omega Indexer and many more. Source: https://ift.tt/9CQAk8J Wiki: https://ift.tt/0Atnkfw https://ift.tt/9CQAk8J September 26, 2024 at 06:46AM
Show HN: Httpdbg – A tool to trace the HTTP requests sent by your Python code Hi, I created httpdbg, a tool for Python developers to easily debug HTTP(S) client requests in Python programs. I developed it because I needed a tool that could help me trace the HTTP requests sent by my tests back to the corresponding methods in our API client. The goal of this tool is to simplify the debugging process, so I designed it to be as simple as possible. It requires no external dependencies, no setup, no superuser privileges, and no code modifications. I'm sharing it with you today because I use it regularly, and it seems like others have found it useful too—so it might be helpful for you as well. Hope you will like it. cle Source: https://ift.tt/0gH4fXc Documentation: https://ift.tt/XViB4Jx A blog post on a use case: https://ift.tt/jp4WwqT... https://ift.tt/0gH4fXc September 26, 2024 at 12:48AM
Show HN: Agentic AI coding tool that's easy to use, transparent and customizable Hey HN, I’d like to share a new agentic AI coding assistant I’ve been working on called OppyDev. Rather than trying to build something that will write the code for you this is aiming for something more collaborative. You can follow along as the AI makes updates and add your own changes on top. We’ve made the inner workings of the agent accessible to the user by outputting its workflow into the UI and allowing most actions to be triggered manually. You can also use the AI to create your own custom plugins to enhance the existing functionality. Here are some videos available demonstrating its usage: - https://www.youtube.com/watch?v=g9FrwVOHTdE&t=187s - https://www.youtube.com/watch?v=6JlQwnYn7RY&t=119s - https://ift.tt/pU68lyK The current version has been released in beta for PC and Mac as an electron app, with a VSCode plugin in the works. You can see a demo video and the full list of our feature set here, https://oppydev.ai . There is more about building plugins at https://ift.tt/YLwtx5y . You can get started for free with your own Pinecone and OpenAI API keys or you can message me for a promo code to get started with free credit. We're looking for feedback so we can improve the tool. We’re also offering free inference costs and support for several small teams that are interested in using the tool and providing feedback on an ongoing basis. Contact me at: https://ift.tt/RjdUz0L https://ift.tt/LK5XclQ https://ift.tt/7IS6mLJ September 25, 2024 at 01:17AM
Show HN: GitHub Repo Visualizer Using D3 I built this as part of my quest to properly learn data visualization. The code is the easy part! Some lessons learned: - personal verification of the the general truth that pie charts are tough! and the returns are not great for the effort due to people's difficulties perceiving angles - may not use "vanilla" d3 with no React. was difficult to adapt for mobile - the GitHub API provides fairly standardized responses so building dynamic charts wasn't too bad. But when working with streaming data (say Kafka) I can see this getting interesting... schema registry should help but creating a view into the data with a lookback would be interesting with d3, done it with altair before. https://ift.tt/vrRLJmU September 24, 2024 at 11:17PM
Show HN: I Wrote a Book on Java https://ift.tt/1oV5usg... This book is a distillation of everything I’ve learned about what effective development looks like in Java (so far!). It's about how to organize programs around data "as plain data" and the surprisingly benefits that emerge when we do. Programs that are built around the data they manage tend to be simpler, smaller, and significantly easier understand. Java has changed radically over the last several years. It has picked up all kinds of new language features which support data oriented programming (records, pattern matching, `with` expressions, sum and product types). However, this is not a book about tools. No amount of studying a screw-driver will teach you how to build a house. This book focuses on house building. We'll pick out a plot of land, lay a foundation, and build upon it house that can weather any storm. DoP is based around a very simple idea, and one people have been rediscovering since the dawn of computing, "representation is the essence of programming." When we do a really good job of capturing the data in our domain, the rest of the system tends to fall into place in a way which can feel like it’s writing itself. That's my elevator pitch! I hope you check it out. I'd love to hear your feedback. You can get 50% off (thru October 9th) with code `mlkiehl` https://ift.tt/1oV5usg... September 24, 2024 at 12:26AM
Show HN: Hoodmaps Meets 4chan I made this as a pair programming exercise with o1-preview. o1 did most of the heavy lifting, through high level prompts, but eventually I needed to diverge from it to get to completion. My initial prompt was: --- I'm making a web app: It's like 4chan, but each thread starts with a gps coordinate. People can only post text and images. I want to build it with flask and sqlalchemy. I want just the backend and data model that supports these threads, anonymous users, and users that can register a permanent username. Write as much of it as you can for me. --- This produced a fully scaffolded Flask app running on sqlite. After that, it took a few (casual) evenings of iteration to dockerize everything, implement postgres instead of sqlite, and use a bucket for storage. On my own, this would have easily taken me 10x the time. https://mapchan.com September 23, 2024 at 06:39PM
Show HN: I built a tool that helps with user onboarding and app understanding Hello HN, In April 2024, I launched my first app, in which people faced difficulties navigating it. Therefore, something had to be done in that regard. My first idea was to make a blog, in which I could present the app’s features so people would get a better understanding of it. However, that did not work, as people simply did not have the time to read all of the long and detailed instructions on how everything works. As this first experiment failed, I decided to add Knowledge Bases, where I took the content of the blogs and basically made them shorter so it would be easier to follow. However, users had difficulties following the instructions as they were simply too boring. It was still a difficult task for the user, as they had to remember each step, and if they forgot something along the way, they had to go back to the Knowledge Base and redo all the steps. At this point, I decided to step things up by creating video tutorials for each of the previous posts I had created. At that time, in my opinion, that had to be the end of the issue: I was literally showing the user what buttons to click step by step. However, despite it being an improvement, it did not solve the problem. People still had issues with finding the same buttons, especially if they were using different devices such as tablets or phones. I realized the solution was close; the only issue now was that the tutorials were hard to follow on different devices. I got tired of googling solutions and decided to create my own tool. The first idea was to directly mark the elements in the code and display pop-ups for each of them. As you may have thought, this was a tedious task because I needed to find the specific elements in my codebase and insert some attributes on them, not to mention that if I needed to use the same element for two different tutorials, I had to mark them differently. That said, I came up with the idea of creating a tool where you had a “Record” button, and once you start a recording, you just click the elements you want to highlight and type a description. With this, I finally had the solution to my problem. It was easy for me to set up, and it was just as easy for the user to follow. After seeing the results, I decided to make a standalone tool that could potentially help others. This is how Tooltip Tutorial was born. It’s very simple to use, it saves a lot of time, and right now it also has comprehensive analytics, which will help with optimizing the onboarding process for your apps. I am eager to see your opinion on this, and if there is anything I can do better, please let me know, and I will do my best to meet those requirements. Thank you! https://ift.tt/uVxFNkm September 23, 2024 at 12:50AM
Show HN: LyricFluent – Learn Languages with Music (Mainly Spanish) My motivation for this product was to 'scratch my own itch'. I speak a few languages, have been learning a few more, and have always enjoyed listening to music in my target language. After failing to find my ideal solution, I built my own. I also wanted to give myself this challenge so that I could become a better full-stack software developer. I built the website with React & Next JS, mobile apps with React Native, and the backend with Node JS. Hosted on DigitalOcean. I worked on this for 4 years (it has been a while), but most of this time was part-time while working another job for income. The URL linked here has free lessons with no account required. Main landing page: https://lyricfluent.com Feedback welcome! https://ift.tt/wsczgNm September 22, 2024 at 06:53PM
Show HN: I built a chatbot to converse with 3M SaaS product reviews Hey HN, I'm Andrei, and together with my co-founder Roman have been working on a tool called Reviewradar. Working with startups, doing interviews and sifting through endless software reviews are daunting tasks. So, we built a chatbot that lets you chat with over 3 million reviews from more than 100K SaaS products. With Reviewradar, you can ask questions like: - create a comprehensive SWOT analysis for both Notion and Obsidian - give me negative feedback and complaints you have about Postmark - summarise the reviews you have on products in the OCR category I would love to get your feedback on it. Check it out here: https://reviewradar.ai Looking forward to your thoughts / suggestions, Andrei and Roman https://reviewradar.ai/ September 22, 2024 at 06:34PM
Show HN: Inbound Email (SMTP) to Webhook Here's my first (hopefully of many) open source release. A minimal script to receive emails via SMTP, parse content (including headers), store attachments in Amazon S3, and forward email content to a webhook. I use it to power DMARC report storage and email content testing. Some of the big email API providers have inbound APIs but costs can rack up fast if you're using them at scale. Hence why I built this. https://ift.tt/2awnoUb Features - SMTP server to receive emails concurrently - Parses incoming emails using mailparser - Uploads attachments to Amazon S3 - Forwards parsed email content to a specified webhook - Configurable via environment variables - Handles large attachments gracefully - Queue system for processing multiple emails and webhook requests simultaneously https://ift.tt/2awnoUb September 22, 2024 at 11:49AM
Show HN: PDF to MD by LLMs – Extract Text/Tables/Image Descriptives by GPT4o I've developed a Python API service that uses GPT-4o for OCR on PDFs. It features parallel processing and batch handling for improved performance. Not only does it convert PDF to markdown, but it also describes the images within the PDF using captions like `[Image: This picture shows 4 people waving]`. In testing with NASA's Apollo 17 flight documents, it successfully converted complex, multi-oriented pages into well-structured Markdown. The project is open-source and available on GitHub. Feedback is welcome. https://ift.tt/KiFaz4T September 22, 2024 at 07:35AM
Show HN: Turn your dreams into reality with this life planner app Hi, this is Ygee, and I'd like to share the story of Y-Pod. Most people believe their big dreams are impossible to achieve, so they settle for less—not because they can’t, but because they feel overwhelmed and disconnected from that distant future. I felt the same way. It was a struggle juggling multiple apps like notes, to-dos, and calendars to plan my goals. I realized a dedicated app to organize my life goals would not only help me but also allow me to showcase my technical and creative skills. That’s how Y-Pod was born — to bridge the gap between big lifelong dreams and small daily actions, empowering users to turn their dreams into reality. My hope is that it will help others live their dream lives and, in doing so, make the world a better place. https://ift.tt/6eF2YUC September 21, 2024 at 06:24AM
Show HN: Open-source text classification CLI – train models with no labeled data aiq lets you use a single CLI command to auto-label text data with a language model API & train an efficient classifier on the labels. Training and inference are super fast (both run on CPU). https://ift.tt/QHNRcxY September 21, 2024 at 12:52AM
Show HN: EloqKV – Scalable distributed ACID key-value database with Redis API We're thrilled to unveil EloqKV, a lightning-fast distributed key-value store with a Redis-compatible API. Built on a new database architecture called the Data Substrate, EloqKV brings significant innovations to database design. Here’s the unique features that makes it stand out: - Flexible Deployment: Run it as a single-node in-memory KV cache, a larger-than-memory database or scale to a highly available, distributed transactional database with ease. - High Performance: Achieves performance levels comparable to top in-memory databases like Redis and DragonflyDB, while significantly outperforming durable KV stores like KVRocks. - Full ACID Transactions: Ensures complete transactional integrity, even in distributed environments. - Independent Resource Scaling: Scale CPU, memory, storage, and logging resources independently to meet your needs. We’d love to hear your thoughts and feedback! https://ift.tt/PtwZz65 September 19, 2024 at 05:33PM
Show HN: An Elliptic Curve-Based Secure Chat, Written Using Rust and Protobuf This is a project I made to learn more about crypto. I've also written a [post]( https://ift.tt/I5neNwp ) that explains all the crypto concepts and algorithms used in the project, and how I implemented them. I'd be very glad for any feedback about both the code and the post :) https://ift.tt/iLwqfOV September 19, 2024 at 11:52PM
Show HN: I Made TabDock–Organize Tabs,Kbar,and Searchbar Commands in the Browser Hey HN! I'm Angel, a recent grad who often juggles multiple projects at once. This leads to a clutter of unrelated tabs in my browser, making it hard to stay focused and organized. To solve this, I built TabDock, a tab manager that helps me stay organized by splitting my work into different windows. Here’s what TabDock offers: - New Tab Dashboard: A clean new tab page with a built-in calendar for quick scheduling. - Kbar Integration: Quickly search through open tabs or other tools. - Search Bar Commands: I’ve added customizable commands to the search bar because they’re cool and useful for opening tabs. I’d love to hear any feedback you have or ways to improve it! Thanks for checking it out! https://ift.tt/qYi3FlR https://ift.tt/qYi3FlR September 19, 2024 at 11:50PM
Show HN: Bot or Not? AI voices vs. humans My cofounder and I were playing with AI voices from different vendors and were shocked at how good the results were. We decided to make a game so you can test your own ability to tell humans from AI voices. This has helped us narrow down which voices we'd like to use for an upcoming project—perhaps you will find it useful and/or interesting too! Would love to hear any feedback you have! https://bot.unison.fm/ September 19, 2024 at 12:27AM
Show HN: GitNotebooks - Jupyter Notebook Review Tool Hey HN, would love some feedback on our Jupyter Notebook review tool! We help data science teams check one another's work and share knowledge that's stored in Jupyter Notebooks. Currently, we integrate with GitHub, allowing for review comment submissions, pull request approval, comments on markdown and code cells that sync with GitHub. Happy to answer any questions! https://ift.tt/GwHP6vh September 18, 2024 at 11:27PM
Show HN: I open-sourced my HN comment/reply alerts service 2 months ago I launched a service that sends email alerts when you get a new comment/reply on Hacker News. I've only been able to acquire 6 users since releasing it (°ー°〃) I tried promoting it here (duuh), Reddit, and LinkedIn. Optimized it for SEO (ranking 3rd on my target keywords). All with limited success. So I've decided to open-source it! Links to... - frontend: https://ift.tt/bjfNiAW... - backend: https://ift.tt/hKA8ads I'm still using it for myself, since there's no service that sends alerts when you get a new comment in one of your posts. But I won't invest any more time in it. Welcome any questions/comments! https://ift.tt/hKA8ads September 18, 2024 at 12:59AM
Show HN: Build event-driven workflows with Python async functions I noticed the new feature of llama-index: the event-driven workflow[1], which allows for building agents based on LLMs. I think the concept of an event-driven workflow is cool, but I don't find the way llama-index implements it to be intuitive. Therefore, I wrote a simple prototype to demonstrate what I believe is the correct approach. It supports the following features with just a few lines of code: - Multi-Recv: it supports an event to listen on a group of other events(or self), and triggered only when the group of events produced the results - Parallel Execution: using python event loop - Dynamic dispatch: it's a cool feature I think. You can use signals like goto or abort to change the event's return behaviors. [1]: https://ift.tt/sXB67PU... https://ift.tt/Ipm2E5P September 17, 2024 at 09:16PM
Show HN: Kty – The Terminal for Kubernetes kty is an SSH server written in rust that provides a TUI-based dashboard mapping k8s concepts onto SSH. It relies on OpenID providers for identity so that you don't need to introduce anything special to your stack. With SSH in place, there are some fun things you can do which are difficult otherwise: - Forward traffic from a service on the cluster to your local laptop using `ssh -R`. - Forward traffic from a local port to the cluster using `ssh -L`. - Use `scp` or any sftp client natively to copy files in and out of containers. - Shell into a container from anywhere that SSH is supported. https://kty.dev September 16, 2024 at 10:16PM
Show HN: AI Chatbot App with Pay-per-Use Access to Multiple Models I've built an AI chatbot app that provides pay-per-use access to multiple high-end AI models from different providers. It's designed for developers and hobbyists who need occasional access to powerful AI models without monthly subscriptions. Features: - Access to models from OpenAI, Anthropic, and others - Pay-per-use credit system - No monthly fees or commitments - Suitable for both hobby projects and professional use I'd love to get your feedback and answer any questions! https://ift.tt/HuDFacd September 16, 2024 at 11:22PM
Show HN: Preppr – a modern prepping/survivalist iOS app Hi, this is my first iOS app and built it for the prepping community as I saw all their existing apps looked like they were from 2010 :) It's 80-90% free (unless you are a power user e.g. create a lot of AI generated items, then there is a cheap pro plan). Enjoy! https://ift.tt/jzfa823 September 14, 2024 at 11:43PM
Show HN: Identify fake news, disinformation and clickbaits with Stampira Stampira is a community-driven search engine that relies on crowd sourced data to filter the search engine result page (serp) from things such as fake news, fake products, misinformation, disinformation, Clickbaits, scams etc by using dedicated stamps that are diagonally overlayed on the snippet of each search result. Stampira users can reward genuine websites that publish high quality articles with stamps such as: Well Researched, Well Written, Educative, Informative etc. Users can also revolt against websites that publish biased articles with stamps like: One-sided, Exaggerated, Misleading etc; Users can also flag AI written articles with the "AI-Generated" stamp. Registered users can select from a variety of stamps that they can use after visiting a website and nonregistered users can still search the web but can't stamp. Currently any website that has the "gov" domain extension can't be stamped, this measure is taken to protect government websites from vandalism and as such will have the "Protected" stamp diagonally overlayed on the snippet by default. The site will be updated on a daily basis as we begin to gather data on how users interact with the site and how they can potentially abuse it. In some cases we may hide the snippet on the serp to encourage users to read before they stamp. https://ift.tt/y4ezL6V September 14, 2024 at 03:25AM
Show HN: A YouTube videos course generator REPOST! (to get more feedback from HN) An exciting aspect of what I'm working on is that users can create their own playlists. For example, you can provide a description of what you're learning, such as a lesson title, and the pipeline will create a youtube playlist. Since the content is highly customized, it will be relevant and high-quality, matching current lessons or lectures in school or university. I'm trying to find a product structure that users will find valuable and effective. The web app is completely free to use right now as I figure out the best model. https://ift.tt/RynlgG0 September 12, 2024 at 10:35PM
Show HN: Pax – cross-platform GUIs with an integrated design tool Hey HN! We are Zack, Warfa, and Sam. After some human-years of building Pax, we're excited finally to enter Beta and to invite anyone to build with Pax for the first time. What is Pax? Pax is a tool for building native apps & websites, similar to SwiftUI or Flutter. Pax is driven by a declarative user interface description language that attaches to Rust application logic. Pax itself is built in Rust. Pax ships with an integrated vector design tool. This design tool is a bidirectional view into any Pax codebase: open codebase with designer, make visual changes, edit pax-lang or Rust by hand with any code editor, and continue to switch back and forth between visual and written modes. [0] Unlike most visual builders, Pax Designer has the tools, features, and conventions of a professional vector design tool (like Figma, Illustrator, or Flash.) This foundational goal required careful design of every aspect of Pax, from the grammar through the rendering engine, the layout system, and the standard library. Try it out You can try out Pax Designer right now on your machine and see the changes it makes to code in realtime: https://ift.tt/Jb9mrNv You can also run Pax Designer directly from our website, but you won't be able to see the code sync: https://www.pax.dev/ Pax Designer goes open source Along with this Beta launch we are open-sourcing Pax Designer[1], Pax’s integrated vector design tool — which itself was built 100% in Pax. That makes Pax Designer a solid reference example of Pax in production. What's next? We're working on a fully-featured hosted version of Pax Designer, which will become Pax Pro[4] — a team collaboration tool that enables non-developers to make visual contributions to GitHub repos hand-in-hand with developers. We are also working on Pax JavaScript, bindings that will allow pax-lang to attach to JavaScript/TypeScript application logic as an alternative to Rust. Other features and fixes will be a function of user feedback. Please take it for a spin, build something, and let us know what you think! See a partial list of current features on the GitHub README.[3] Please reach out to our team with any questions, ideas, or feedback! This thread, GitHub issues, or Discord[5] are good ways to get in touch. Pax today in Beta is far from perfect, but we're proud of how far it's come and excited about where it's headed. We hope some folks here will share our excitement, or even join us in our mission to make software creation more creative and accessible to more of humanity. -- [0] How Pax Designer reads & writes code: https://ift.tt/0wqU7xa [1] Source code for Pax Designer, built in Pax: https://ift.tt/6xqKZF2 [2] Get started: https://ift.tt/Jb9mrNv [3] GitHub Repo + Readme: https://ift.tt/3AdgWzu [4] Get early access to Pax Pro: https://ift.tt/NvfKId3... [5] Discord: https://ift.tt/s5PhNz3 https://www.pax.dev/ September 11, 2024 at 10:47PM
Show HN: Tune LLaMa3.1 on Google Cloud TPUs Hey HN, we wanted to share our repo where we fine-tuned Llama 3.1 on Google TPUs. We’re building AI infra to fine-tune and serve LLMs on non-NVIDIA GPUs (TPUs, Trainium, AMD GPUs). The problem: Right now, 90% of LLM workloads run on NVIDIA GPUs, but there are equally powerful and more cost-effective alternatives out there. For example, training and serving Llama 3.1 on Google TPUs is about 30% cheaper than NVIDIA GPUs. But developer tooling for non-NVIDIA chipsets is lacking. We felt this pain ourselves. We initially tried using PyTorch XLA to train Llama 3.1 on TPUs, but it was rough: xla integration with pytorch is clunky, missing libraries (bitsandbytes didn't work), and cryptic HuggingFace errors. We then took a different route and translated Llama 3.1 from PyTorch to JAX. Now, it’s running smoothly on TPUs! We still have challenges ahead, there is no good LoRA library in JAX, but this feels like the right path forward. Here's a demo ( https://ift.tt/UYEKRvD ) of our managed solution. Would love your thoughts on our repo and vision as we keep chugging along! https://ift.tt/8vNXrCc September 11, 2024 at 08:44PM
Show HN: Anthropic React Native Client I recently published a new package on npm that brings the Anthropic APIs to React Native without polyfills. The library supports chat streaming and normal chat completions. The goal of the library is to follow the Node SDK wherever possible while taking advantage of React Native SSE for streaming where the Anthropic Node SDK does not work. Lmk what you think or if this will be useful to you! https://ift.tt/gH8t7Qe September 10, 2024 at 09:30PM
Show HN: Turn Any ArXiv Paper into a 200-Page Prerequisite Reading Book I created this tool over the weekend because, as someone interested in AI and technology, I find many research papers on arXiv fascinating but often incredibly dense and difficult to understand due to the heavy jargon and technical language. I wanted to make these complex topics more accessible to laypeople like myself by converting the topics described in these papers into full-length books that break down key concepts, making cutting-edge research easier to grasp. You can think of it as generating prerequisite reading materials before being able to read the actual paper. Here are some noteworthy arXiv papers you can test this with: - Attention is All You Need: https://ift.tt/haVpYnf - BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding: https://ift.tt/lerK3Th - GPT-3: Language Models are Few-Shot Learners: https://ift.tt/QFfh8xJ - Deep Residual Learning for Image Recognition (ResNet): https://ift.tt/qNhiJSe The tool leverages GPT-4o, Perplexity, and Instructor to analyze and break down the complex concepts within these papers. Keep in mind, it's not built for heavy traffic (current capacity is about 50 books per hour) so if things get busy, it may take a bit longer, but the book will arrive via email eventually! https://ift.tt/liGSrJ6 September 9, 2024 at 11:53PM
Show HN: SocialPredict, MIT Licensed, Easy to Deploy Prediction Market Platform Started this last October with the idea of enabling small communities to deploy their own walled-off prediction market instances. We are being used by Kenyon College and St. Lawrence University. To help please star, fork and suggest fixes, bugs or issues. Please submit a discussion on Github discussions if you would like help spinning up your own. https://ift.tt/GUcWfwX... https://ift.tt/1l89Wyu September 9, 2024 at 10:01PM
Show HN: Ki Editor - the multicursor syntactical editor Hi everyone, I have been developing this editor, Ki, for over a year, and have employed it substantially in all kinds of development (including Ki itself) for at least 3 months. I think it is mostly crystallized, thus I'm happy to share it with you today. Its main strength is first-class multi-cursor and structural (syntax) editing, which is a rare combination in the realm of editors (TUI or GUI alike). Hope you'll enjoy it! https://ift.tt/taIsgxm September 9, 2024 at 06:09AM
Show HN: Dump entire Git repos into a single file for LLM prompts Hey! I wanted to share a tool I've been working on. It's still very early and a work in progress, but I've found it incredibly helpful when working with Claude and OpenAI's models. What it does: I created a Python script that dumps your entire Git repository into a single file. This makes it much easier to use with Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) systems. Key Features: - Respects .gitignore patterns - Generates a tree-like directory structure - Includes file contents for all non-excluded files - Customizable file type filtering Why I find it useful for LLM/RAG: - Full Context: It gives LLMs a complete picture of my project structure and implementation details. - RAG-Ready: The dumped content serves as a great knowledge base for retrieval-augmented generation. - Better Code Suggestions: LLMs seem to understand my project better and provide more accurate suggestions. - Debugging Aid: When I ask for help with bugs, I can provide the full context easily. How to use it: Example: python dump.py /path/to/your/repo output.txt .gitignore py js tsx Again, it's still a work in progress, but I've found it really helpful in my workflow with AI coding assistants (Claude/Openai). I'd love to hear your thoughts, suggestions, or if anyone else finds this useful! https://ift.tt/fqjBGAH P.S. If anyone wants to contribute or has ideas for improvement, I'm all ears! September 9, 2024 at 01:38AM
Show HN: Magic List – Super Simple Todo List - After trying out many TODO lists wanted something that was super simple to use and share with family and friends - Runs entirely on the browser. No servers involved other than for downloading HTML/CSS/JavaScript - Sharing adds the contents of the list as URL query parameter which helps avoid need for a server. But limits size of lists that can be shared to around 2KB. - Comments, Feedback, Feature Requests Welcome! https://ift.tt/FrUQ41B September 7, 2024 at 11:24PM
Show HN: Museum Music – generate period-appropriate playlists from artwork This is Museum Music, an app to help experience art with era-appropriate music - photograph an art piece, and it will create a unique Spotify playlist that complements its era and style. We built this from our own experiences visiting museums and thoughts of enhancing the experience with music. It's built with GPT Vision and the Spotify API. Although the initial focus was on art-inspired music, it works surprisingly well on other, day-to-day photos too! https://ift.tt/JX5dmzu September 7, 2024 at 01:52AM
Show HN: Draw from your webcam with pinch ui try it out! (on your computer, though phone teeeechnically works) Uses webcam ("pinch ui") to draw! Originally used as a tool to help kids learn to write, but who knows what else? (maybe a better draw during zoom meetings? ...) this is very early stage (there may be some bugs/glitches), so please let me know of them, and let me know what you think! https://ift.tt/0J8ABmd September 7, 2024 at 02:57AM
Show HN: Nomadic – Minimize RAG Hallucinations with 1 Hyperparameter Experiment Hey HN! Mustafa, Lizzie, and Varun here from NomadicML ( https://nomadicml.com ). We’re excited to show you Nomadic ( https://ift.tt/2PLhQYf ): a platform focused on parameter search to continuously optimize AI systems. Here’s a simple demo notebook where you get the best-performing, statistically significant configurations for your RAG — and improve hallucination metrics by 4X in just 5 minutes — with a single Nomadic experiment: https://ift.tt/jGynMLs Our lightweight library is now live on PyPI (`pip install nomadic`). Try one of the README examples :) Input your model, define an evaluation metric, specify the dataset, and choose which parameters to test. Nomadic emerged from our frustration with existing HPO (hyperparameter optimization) solutions. We heard over and over that for the sake of deploying fast, folks resort to setting HPs through a single, expensive grid search or better yet, intuition-based “vibes”. From fine-tuning to inference, small tweaks to HPs can have a huge impact on performance. We wanted a tool to make that “drunken wander” systematic, quick, and interpretable. So we started building Nomadic - our goal is to create the best parameter search platform out there for your ML systems to keep your hyperparameters, prompts, and all aspects of your AI system production-grade. We started aggregating top parameter search techniques from popular tools and research (Bayesian Optimizations, cost-frugal flavors). Among us: Built Lyft’s driver earnings platform, automated Snowflake’s just-in-time compute resource allocation, became a finalist for the INFORMS Wagner Prize (top prize in industrial optimization), and developed a fintech fraud screening system for half a million consumers. You might say we love optimization. If you’re building AI agents / applications across LLM safety, fintech, support, or especially compound AI systems (multiple components > monolithic models), and want to deeply understand your ML system’s best levers to boost performance as it scales - get in touch. Nomadic is being actively developed. Up next: Supporting text-to-SQL pipelines (TAG) and a Workspace UI (preview it at https://ift.tt/MEQDK2l ). We’re eager to hear honest feedback, likes, dislikes, feature requests, you name it. If you’re also a optimization junkie, we’d love for you to join our community here https://ift.tt/JVhiCzy September 5, 2024 at 11:44PM
Show HN: DevMuse – App to bond over music and code I've found there's a disconnect devs can feel when they are remote only, and one of the ways people have connected since the beginning of humanity has been through music. So I thought it would be interesting to build off these concepts and build a tool that brings a sense of connection to devs across the world. I don't have it fully working just yet. Still have some kinks to work out. What do you think? https://ift.tt/BNQTlr7 September 6, 2024 at 01:39AM
Show HN: Starling – A Cross-Platform File Management System with Version Control Hey HN, I’m excited to introduce you to Starling, a new way to manage files and folders across platforms. > Why Starling? There are many collaboration tools on the market. And they work great. But they work great in a fixed setting: Like a business decides to roll out system XYZ. Or a group of friends decide to use cloud ABC. But what if I often change projects and collaborators? What if for my profession I work primarily with native applications? This means I’ll send and receive data through diverse and ad-hoc channels: Put something in the cloud and send a link. Receive a link to something in the cloud. Receive an email with attachment. Etc. What app is out there that puts me in control of this, without making demands of others? That is Starling: Starling is a file management system designed to keep your dynamic data organized, prevent version conflicts, and loss. Here are some of its features: - Total Version Control: Track and manage file versions out of the box, so you never end up with wrong files or ambiguous lists of filenames. See immediately who did what, when; and you decide what is the actual final version. - Cross-Platform and Offline Compatibility: Works with Apple, Microsoft, and Linux desktops, enabling seamless collaboration regardless of the software ecosystem. No account or login needed: work whenever you want, wherever you want. No cloud back-end, no lock in. - Data Insight & Organization: Starling offers a complete overview of your data, dynamically tracking relationships and data. Starling offers you data time-traveling, instead of “time-flattened” endless lists of files. Easily look what you were working on e.g. last December. - Privacy & Security: Your data stays yours. Starling operates without centralization, putting data privacy and control first. Starling’s core value is that it wants to know as little as possible about you. Again, no login or account needed. Starling works offline. > What holds the future? The software is stable but we’re a bootstrapped startup, and the software would greatly benefit from a more comprehensive team. So getting this software into the market is part of our wish to grow so we can further improve our app. In terms of actual features, we have a peer-to-peer network in alpha which would allow you to seamlessly work together with anybody who also has Starling, right from your desktop. Let me explain: The core idea of the software is to put you back in charge of data. Too much time is lost digging up data because there’s no tool that says “hey, do your thing, I’ll keep an eye on things for you”. My personal, reliable, data assistant. Next step would be that it’d be nice if I could interface seamlessly with those collaborators who also have Starling, and simply designate certain data we work on together. No copying of files needed, no dragging of folders into the cloud, or exporting to doc and then downloading. Just simple and easy file management for the busy individual right from your desktop. > Looking for Feedback Starling is in active development and would love feedback from the HN community. If you have thoughts on file management, version control, or if you just want to try out something new, we'd be grateful for your insights. We’d love for you to try our software, and would be honored if it integrates into your day-to-day work. The trial is no strings attached: your first fifty files are free so there’s no time limit. Also, we are looking for a limited number of users who’d like to actively engage with us on making the software better. If this is you, simply drop us a line by replying on the trial confirmation email you’ll receive after finalizing app setup. In return we gladly offer you a year of our Pro plan for free. Happy to answer any questions and looking forward to your feedback! Thank you for your attention! Willem, Founder https://ift.tt/fb3RW29 September 5, 2024 at 05:21PM
Show HN: Local Bookmark Storage and Archive I decided I needed a reliable backup of my bookmarks, and an archive of those pages for future reference in case they get taken down. I paid for a managed online service that did this for a while, but I was unsatisfied with the archiving reliability. This is pretty much just a 200 line python script, but it does the trick, and now I don't have to worry about losing a copy of a web page that I want to refer to in the future. https://ift.tt/XV4WPES September 5, 2024 at 01:29AM
Show HN: Open-source Hacker News for Android client We believe the best software is built when devs seriously use the thing they're building. We're a mobile dev tools company (emergetools.com), but we don't actually build & ship "apps"... until now. Excited to share our first open-source app: Hacker News for Android Any feedback/PRs very welcome on the repo. We also wrote about why we built and open-sourced this HN app here: https://ift.tt/n5P6oHW... https://ift.tt/dWHm4wh September 5, 2024 at 01:02AM
Show HN: ASim – generate functional mobile apps instantly on your phone Hi HN! Daniel from YC S21 here, launching a project called aSim, which lets people generate/simulate usable native apps (called Sims) instantly from your phone. Describe an app you want and aSim will generate it. Then, edit and refine it to better suit your needs. Sims are also shareable via links, and basic app functionality is also exposed through web (though mobile is much more feature complete). A couple of my favorite Sims so far: - FridgeChef, recipes from what's in your fridge: https://asim.sh/s/419 - wordle: https://asim.sh/s/412 - HIIT workout randomizer/generator: https://asim.sh/s/418 - Pomodoro timer: https://asim.sh/s/232 Would love feedback around the experience and additional functionality you'd like surfaced! https://asim.sh/ September 4, 2024 at 11:03PM
Show HN: An open-source implementation of AlphaFold3 Hi HN - we’re the founders of Ligo Biosciences and are excited to share an open-source implementation of AlphaFold3, the frontier model for protein structure prediction. Google DeepMind and their new startup Isomorphic Labs, are expanding into drug discovery. They developed AlphaFold3 as their model to accelerate drug discovery and create demand from big pharma. They already signed Novartis and Eli Lilly for $3 billion - Google’s becoming a pharma company! ( https://ift.tt/Yg5UFwN... ) AlphaFold3 is a biomolecular structure prediction model that can do three main things: (1) Predict the structure of proteins; (2) Predict the structure of drug-protein interactions; (3) Predict nucleic acid - protein complex structure. AlphaFold3 is incredibly important for science because it vastly accelerates the mapping of protein structures. It takes one PhD student their entire PhD to do one structure. With AlphaFold3, you get a prediction in minutes on par with experimental accuracy. There’s just one problem: when DeepMind published AlphaFold3 in May ( https://ift.tt/N7y12Wv ), there was no code. This brought up questions about reproducibility ( https://ift.tt/atxsMvN ) as well as complaints from the scientific community ( https://ift.tt/R59KFOP... ). AlphaFold3 is a fundamental advance in structure modeling technology that the entire biotech industry deserves to be able to reap the benefits from. Its applications are vast, including: - CRISPR gene editing technologies, where scientists can see exactly how the DNA interacts with the scissor Cas protein; - Cancer research - predicting how a potential drug binds to the cancer target. One of the highlights in DeepMind’s paper is the prediction of a clinical KRAS inhibitor in complex with its target. - Antibody / nanobody to target predictions. AlphaFold3 improves accuracy on this class of molecules 2 fold compared to the next best tool. Unfortunately, no companies can use it since it is under a non-commercial license! Today we are releasing the full model trained on single chain proteins (capability 1 above), with the other two capabilities to be trained and released soon. We also include the training code. Weights will be released once training and benchmarking is complete. We wanted this to be truly open source so we used the Apache 2.0 license. Deepmind published the full structure of the model, along with each components’ pseudocode in their paper. We translated this fully into PyTorch, which required more reverse engineering than we thought! When building the initial version, we discovered multiple issues in DeepMind’s paper that would interfere with the training - we think the deep learning community might find these especially interesting. (Diffusion folks, we would love feedback on this!) These include: - MSE loss scaling differs from Karras et al. (2022). The weighting provided in the paper does not downweigh the loss at high noise levels. - Omission of residual layers in the paper - we add these back and see benefits in gradient flow and convergence. Anyone have any idea why Deepmind may have omitted the residual connections in the DiT blocks? - The MSA module, in its current form, has dead layers. The last pair weighted averaging and transition layers cannot contribute to the pair representation, hence no grads. We swap the order to the one in the ExtraMsaStack in AlphaFold2. An alternative solution would be to use weight sharing, but whether this is done is ambiguous in the paper. More about those issues here: https://ift.tt/MnK7hIu How this came about: we are building Ligo (YC S24), where we are using ideas from AlphaFold3 for enzyme design. We thought open sourcing it was a nice side quest to benefit the community. For those on Twitter, there was a good thread a few days ago that has more information: https://twitter.com/ArdaGoreci/status/1830744265007480934 . A few shoutouts: A huge thanks to OpenFold for pioneering the previous open source implementation of AlphaFold We did a lot of our early prototyping with proteinFlow developed by Lisa at AdaptyvBio we also look forward to partnering with them to bring you the next versions! We are also partnering with Basecamp Research to supply this model with the best sequence data known to science. Matthew Clark ( https://batisio.co.uk ) for his amazing animations! We’re around to answer questions and look forward to hearing from you! https://ift.tt/MnK7hIu September 4, 2024 at 11:14PM
Show HN: Icebreaking AI. A free tool to help you find close friends Hello, everyone! My name is Alex, and I'm an expat in a new country. Two years ago, I moved to Germany where I didn’t know anyone. Starting the job, I realized that becoming close friends with colleagues can be quite challenging and slow. The feeling of loneliness was gradually growing. To change that, I started attending various meetups based on my interests, hoping to make new friends. However, I often found myself answering the same generic questions that didn’t really help me understand if someone shared my values and interests. As a result, lost energy and 0 friends. SO - That’s when I began using icebreaking questions in smaller groups. Surprisingly, these questions quickly changed any conversation to a deeper, more engaging and meaningful level. I was able to identify people I genuinely connected with and started spending more time with them. BUT - preparing and using these questions was a pain. I would Google the top-100, pick a random number, and choose a question. I felt a strong need for a simple „one button“ that does everything for me. AND - After searching and not finding such a solution, I decided to build it myself. I created a vast database of questions across various topics and integrated AI to generate questions on any subject. The idea has resonated well within my communities (university and expat parties), and more people are starting to use it. I’d love to hear your stories with icebreaking questions, your experience, and your feedback so I can make the service even more useful for more people! Check it out and share your thoughts! https://ift.tt/TFCQMYE September 4, 2024 at 01:24AM
Show HN: I made a REST based alternative to GraphQL for PostgreSQL Hi guys! I have had a love-hate relationship with GraphQL for a while and decided to build my own tool as a side project! It's a REST-based API generator that you can use to build and deploy on top of PostgreSQL. I am currently working on a GitHub integration so you can push your API directly to GitHub and deploy on your own infrastructure. The plan is to stabilize the features by December and then to open-source it. It's completely free to use since it's in beta, and I would appreciate any feedback I can get. Thanks: kabir@querydeck.io https://querydeck.io/ September 3, 2024 at 11:15PM
Show HN: A YouTube videos course generator An exciting aspect of what I'm working on is that users can create their own playlists. For example, you can provide an outline of what you're learning, such as notes from your professor or any other structure, and the pipeline will create a course tailored to that. Since the content is highly customized, it will be relevant and high-quality, matching current lessons or lectures in school or university. Initially, the pipeline also generated some teaching content using AI, but due to the risk of AI hallucinations, I'm planning to pivot away from that. The new focus will be on curating YouTube videos relevant to each subject, so the courses will consist entirely of expert-created content from YouTube. This approach avoids copyright issues entirely. I'm still in the process of updating the code to implement these changes. The reason I'm posting this is to get feedback on how it sounds to you as learner. I'd love to hear your thoughts, does this solve a problem for you, would you be likely to use it if it just works ? I'm trying to find a product structure that users will find valuable and effective. The web app is completely free to use right now as I figure out the best model. Thanks! https://coursely.ai September 2, 2024 at 11:00PM
Show HN: Ban 'Any updates' emails – this is super-efficient client communication Looking for 100 beta users to use RSP for free (in return for great feedback!) I've made a Monday/ Google Sheets hybrid - making it really easy to share structured data with clients, in real-time, with no extra effort from you. This eliminates clients being in the dark or sending multiple emails per task. Project management tools are usually black boxes, where the team only communicates with clients via email threads (in the plural). Instead, with RSP, you choose at the column level what data you want to share with clients and what you want them to edit and not edit, and also have a side-by-side messenger for each task. Client communication is instant, real-time, and super-efficient. https://ift.tt/4p0FBxS September 2, 2024 at 02:40PM
Show HN: Bin-graph: Visualize binary files This program provides a simple way of visualizing the different regions of a binary file. Written in C, depends only on libpng. Currently (commit 1dd42e3) it is able to generate PNG images that represent various aspects of the binary: - Grayscale: Byte values, 00..FF. - Ascii: Printability of each byte. - Entropy: Of a "block", changed with --block-size. - Histogram: Bar graph of the byte frequencies. - Bigrams: Each point is determined by a pair of bytes. - Dotplot: Measure self-similarity. Image width/height is N^2. In the future, I plan on adding an SDL version that allows the user to view a section of the file interactively (sections are currently supported with --offset-start and --offset-end). More information on the README. https://ift.tt/td7UXyK September 1, 2024 at 08:00PM
Show HN: I made an app to discover where to stream movies and TV shows in a week I set myself a challenge to make a simple and clear website to discover where to watch movies and TV series within a week. This was my second project with my now favorite tech stack: Next.js, TypeScript, Tailwind and ShadCN UI. https://ift.tt/ElTx62h September 1, 2024 at 04:59PM
Show HN: I'm 17 and built a startup to find or sell parking spots Hey HN! I'm Dris, I'm 17, and for the past few months, I've been solo-developing Parkt, an app that lets you quickly reserve or sell parking spots. Parking sucks - it's expensive, hard to find, and you usually overpay or get a ticket. With Parkt, you can reserve parking in 2 taps, or list your unused space in under 5 minutes. It's now free to download on iOS (with Android coming soon)! You can also check out a demo on the site if you're interested. If you know someone who might use Parkt, sharing the app would be super helpful! Feedback is also appreciated. https://parkt.app/ September 1, 2024 at 12:23AM
संदेश
