Wednesday, July 31, 2024
Show HN: Shimmr: Free iOS App Uses CIA Findings to Enhance Meditation and Focus https://ift.tt/NR3B7KY
Show HN: Shimmr: Free iOS App Uses CIA Findings to Enhance Meditation and Focus https://ift.tt/sHTLW6o July 31, 2024 at 10:28PM
Tuesday, July 30, 2024
Show HN: Heyya v1.0.0 Elixir and Phoenix LiveView Snapshot Testing Library https://ift.tt/dRpsHco
Show HN: Heyya v1.0.0 Elixir and Phoenix LiveView Snapshot Testing Library https://ift.tt/EehcLAw July 30, 2024 at 09:15PM
Monday, July 29, 2024
Show HN: Chrome Extension to Open Google Maps Locations in Apple Maps https://ift.tt/u0rbyTX
Show HN: Chrome Extension to Open Google Maps Locations in Apple Maps Pretty simple extension that displays a prompt notification on Google Maps to open the address you’re viewing in Apple Maps. https://ift.tt/PDn0xXB July 29, 2024 at 09:03PM
Sunday, July 28, 2024
Saturday, July 27, 2024
Show HN: Semantic Grep – A Word2Vec-powered search tool https://ift.tt/yQPf0Lp
Show HN: Semantic Grep – A Word2Vec-powered search tool Much improved new version. Search for words similar to the query. For example, "death" will find "death", "dying", "dead", "killing"... Incredibly useful for exploring large text datasets where exact matches are too restrictive. https://ift.tt/Zxm9VXs July 27, 2024 at 10:02PM
Friday, July 26, 2024
Thursday, July 25, 2024
Show HN: A personalised AI tutor with < 1s voice responses https://news.ycombinator.com/item?id=41057030
Show HN: A personalised AI tutor with < 1s voice responses TLDR: We created a personalised Andrej Karpathy tutor that can response to questions about his Youtube videos in sub 1 second responses (voice-to-voice). We do this using a voice enabled RAG agent. See later in the post for demo link, Github Repo and blog write up. A few weeks ago we released the worlds fastest voice bot, achieving 500ms voice-to-voice response times, including a 200ms delay waiting for a user to stop speaking. After reaching the front page of HN, we thought about how we could take this a step further based on feedback we were getting from the community. Many companies were looking for a way to implement function calling and RAG with voice interfaces while retaining a low enough latency. We couldn’t find many resources about how to do this online that: 1. Allowed us to achieve sub-second voice-to-voice latency 2. Was more flexible than existing solutions. Vapi, Retell, [Bland.ai]( http://Bland.ai ) are too opinionated plus since they just orchestrate API’s which incur network latency at every step. See requirement above 3. The unit economics actually work at scale. So we decided to create a implementation of our own. Process: As we mentioned in our previous release, if you want to achieve response times this low you need to make everything as local as possible. So below was our setup - Local STT: Deepgram model - Local Embedding model: Nomic v1.5 - Local VectorDB: Turso - Local LLM: Llama 3B - Local TTS: Deepgram model From our previous example, the only new components where: - Local Embedding model: We chose Nomic Embed text v1.5 model that gave a processing time of roughly ~200ms - Turso offers local embedded replicas combined with edgeDB’s which meant we were able to achieve 0.01 second read times. Pinecone also gave us good times of 0.043 seconds. The above changes led us to achieve sub 1 second voice-to-voice response times Application: With Andrej Karpathy’s announcement around [Eureka Labs]( https://eurekalabs.ai/ ), a new AI+Education company we thought we would create our very own personalised Andrej tutor. Listen to anyone of his Youtube lectures, as soon as your start specking, the video will pause and he will reply. Once your question has been answered you can then tell him to continue with the lecture and the video will automatically start playing. Demo: https://educationbot.cerebrium.ai/ Blog: https://www.cerebrium.ai/blog/creating-a-realtime-rag-voice-... Github Repo: https://github.com/CerebriumAI/examples/tree/master/19-voice... For demo purposes: - We used OpenAI for GPT-4-mini and embeddings (its cheaper to run on a CPU than GPU’s when running demos at scale. These changes add about ~1 second to the response time - We used Eleven labs to clone his voice to make replies sound more realistic. This adds about 300ms to the response time. The improvements that can be made which we would like the community to contribute to are: - Embed the video screens as well that when you ask certain questions it can show you the relevant lecture slide for the same chuck that it got context from to answer. - Insert the timestamps in the vectorDB timestamps so that if a question will be answered later in the lecture he can let you know This unlocks so many use cases in education, employee training, sales etc that it would be great to see what the community builds! https://educationbot.cerebrium.ai/ July 24, 2024 at 05:41PM
Wednesday, July 24, 2024
Tuesday, July 23, 2024
Show HN: React Native Finisher Kit That'll Automate All the Complex Stuff https://ift.tt/1O5pJqQ
Show HN: React Native Finisher Kit That'll Automate All the Complex Stuff Hey HN, I've been automating mobile builds since 2017, publishing a course on mobile automation back then. Over the years, I’ve configured Mobile CI/CD for various companies using Bitrise, Codemagic, GitHub Actions – over and over again. To simplify this repetitive process, I’ve created a React Native "finisher kit" that automates Mobile CI/CD setup. Here’s what it offers: - Initial setup with monorepo support, multiple environments, unified configuration, testing set up, features like Push or iOS Live Activities… - Easy app creation for App Stores + Fully automated code signing, building, deployment, - Single place for configuration (AWS), reusable for local development, builds and deployment either locally or in the cloud, - A seamless system to manage configurations for Apple, Fastlane, Google, and more… It allows you to create your Mobile CI/CD with any of these providers with a single CLI command and keep it up-to-date. Switch between providers just by changing the config. It supports both React Native CLI projects and Expo. I built it to save time and reduce the complexity of mobile app development, hoping it will be as helpful for you as it has been for me. I’d love to get your feedback and thoughts. Thanks! Igor https://ift.tt/l7vewng July 23, 2024 at 09:42PM
Monday, July 22, 2024
Show HN: Google URL Shortener Import Tool https://ift.tt/oNPT30b
Show HN: Google URL Shortener Import Tool Hey HN, since Google announced a few days ago that their short links will stop working next year[1], I created a small tool that allows you to import batches of goo.gl short links into the URL Shortener I maintain (Short Menu) for free. After the import you'll get a CSV with a mapping between old/new links so you can easily update your content. I figured this could be quite helpful for those who still rely on functioning goo.gl links and speed up the migration a bit. Links: [1] https://ift.tt/t5pfjoV https://ift.tt/CWfFih0 July 22, 2024 at 11:46PM
Sunday, July 21, 2024
Saturday, July 20, 2024
Friday, July 19, 2024
Thursday, July 18, 2024
Show HN: How we leapfrogged traditional vector based RAG with a 'language map' https://ift.tt/TpDrRe7
Show HN: How we leapfrogged traditional vector based RAG with a 'language map' TL;DR: Vector-based RAG performs poorly for many real-world applications like codebase chats, and you should consider 'language maps'. Part of our mission at Mutable.ai is to make it much easier for developers to build and understand software. One of the natural ways to do this is to create a codebase chat, that answer questions about your repo and help you build features. It might seem simple to plug in your codebase into a state-of-the-art LLM, but LLMs have two limitations that make human-level assistance with code difficult: 1. They currently have context windows that are too small to accommodate most codebases, let alone your entire organization's codebases. 2. They need to reason immediately to answer any questions without thinking through the answer "step-by-step." We built a chat sometime a year ago based on keyword retrieval and vector embeddings. No matter how hard we tried, including training our own dedicated embedding model, we could not get the chat to get us good performance. Here is a typical example: https://ift.tt/8p2TjAc... If you ask how to do quantization in llama.cpp the answers were oddly specific and seemed to pull in the wrong context consistently, especially from tests. We could, of course, take countermeasures, but it felt like a losing battle. So we went back to step 1, let’s understand the code, let’s do our homework, and for us, that meant actually putting an understanding of the codebase down in a document — a Wikipedia-style article — called Auto Wiki. The wiki features diagrams and citations to your codebase. Example: https://ift.tt/u96KcGt This wiki is useful in and of itself for onboarding and understanding the business logic of a codebase, but one of the hopes for constructing such a document was that we’d be able to circumvent traditional keyword and vector-based RAG approaches. It turns out using a wiki to find context for an LLM overcomes many of the weaknesses of our previous approach, while still scaling to arbitrarily large codebases: 1. Instead of context retrieval through vectors or keywords, the context is retrieved by looking at the sources that the wiki cites. 2. The answers are based both on the section(s) of the wiki that are relevant AND the content of the actual code that we put into memory — this functions as a “language map” of the codebase. See it in action below for the same query as our old codebase chat: https://ift.tt/8p2TjAc... https://ift.tt/8p2TjAc... The answer cites it sources in both the wiki and the actual code and gives a step by step guide to doing quantization with example code. The quality of the answer is dramatically improved - it is more accurate, relevant, and comprehensive. It turns out language models love being given language and not a bunch of text snippets that are nearby in vector space or that have certain keywords! We find strong performance consistently across codebases of all sizes. The results from the chat are so good they even surprised us a little bit - you should check it out on a codebase of your own, at https://wiki.mutable.ai , which we are happy to do for free for open source code, and starts at just $2/mo/repo for private repos. We are introducing evals demonstrating how much better our chat is with this approach, but were so happy with the results we wanted to share with the whole community. Thank you! https://twitter.com/mutableai/status/1813815706783490055 July 18, 2024 at 10:40PM
Wednesday, July 17, 2024
Show HN: Product Hunt for Music https://ift.tt/hcBwXMZ
Show HN: Product Hunt for Music https://tracklist.it/ July 17, 2024 at 11:31PM
Tuesday, July 16, 2024
Show HN: Contacts: A Microlang for Managing Contacts https://ift.tt/HOD98LJ
Show HN: Contacts: A Microlang for Managing Contacts https://ift.tt/fsaX3R1 July 16, 2024 at 11:38PM
Monday, July 15, 2024
Show HN: TargetJ – New JavaScript framework that can animate anything https://ift.tt/TCtwoNv
Show HN: TargetJ – New JavaScript framework that can animate anything I am excited to introduce to you TargetJ, a new JavaScript framework that can animate anything. I have been working on this project for over two years, driven by the complexity of current UI frameworks. You can find the interactive documentation at www.targetj.io. I hope you find it useful for creating great web experiences. If you have any questions about the framework or want to share your thoughts, please leave a comment below. I’m eager to hear from you! https://ift.tt/urEQiKd July 15, 2024 at 10:36PM
Sunday, July 14, 2024
Saturday, July 13, 2024
Friday, July 12, 2024
Thursday, July 11, 2024
Show HN: Upload your photo and generate crazy YouTube Faces for your thumbnail https://ift.tt/mMIUdWG
Show HN: Upload your photo and generate crazy YouTube Faces for your thumbnail Upload your photo, this AI tool generates hundreds of High-Conversion Youtube faces. Our AI analyzed millions of viral video thumbnails, found the top performing Youtube Faces templates for each niche. Then it can select and generate the best performing youtube faces according to your content. Works for both realistic photos and cartoon photos for faceless channels. https://ift.tt/HvIjoCt July 11, 2024 at 10:57PM
Wednesday, July 10, 2024
Show HN: Sentinel-1 Explorer App Simplifies Access to SAR Imagery https://ift.tt/GfpTXF7
Show HN: Sentinel-1 Explorer App Simplifies Access to SAR Imagery Bringing together the ready-to-use Sentinel-1 RTC imagery from Living Atlas, and core capabilities via the ArcGIS Maps SDK for JavaScript, Sentinel-1 Explorer aims to help democratize SAR imagery for Earth science and observation. https://ift.tt/ebHIz2p July 10, 2024 at 08:46PM
Tuesday, July 9, 2024
Show HN: Oneil, design specification language for rapid system modeling https://ift.tt/2Z9ctk8
Show HN: Oneil, design specification language for rapid system modeling Hi all! I've been doing spacecraft system engineering since grad school and got to keep doing it after school at Care Weather. I tried several approaches for keeping hundreds of napkin math equations and connected simulations up-to-date, but none of them were both reliable and easy to use. To solve that, I made Oneil, a design specification language with clean syntax for specifying and evaluating math describing a complex system. Oneil has been really helpful as we design one satellite iteration after another for ground testing and spaceflight. I'm really interested in hearing feedback and whether anyone finds this useful. If you know of other tools that do the same things, please let me know as well, since I haven't been able to find any. https://ift.tt/avpfy45 July 9, 2024 at 11:30PM
Monday, July 8, 2024
Show HN: WAL Implementation in Golang https://ift.tt/3ODA4fb
Show HN: WAL Implementation in Golang I wrote this simple WAL library in Golang that I use to write data that my kafka producer fails due to errors like Broker going down or some other issue. Took inspiration from etcd/wal https://ift.tt/rxXcZOE July 8, 2024 at 10:50PM
Sunday, July 7, 2024
Show HN: Imageprompt.io – AI-powered and human-curated artworks, logos, photos https://ift.tt/D0tI7J6
Show HN: Imageprompt.io – AI-powered and human-curated artworks, logos, photos ImagePrompt.io is a constantly growing collection of AI-generated content for artists, designers and everyone in need of great images. Includes stock photos for presentations and website, artworks for prints and wallpapers and logos for your next business. https://imageprompt.io/ July 7, 2024 at 11:03PM
Saturday, July 6, 2024
Friday, July 5, 2024
Show HN: Nitrogen Blue Launching from the Forest https://ift.tt/GTQBFfN
Show HN: Nitrogen Blue Launching from the Forest https://ift.tt/eqzY3Dp July 5, 2024 at 10:49PM
Show HN: Foorr – A minimal to-do app with social accountability https://ift.tt/UTqIQlM
Show HN: Foorr – A minimal to-do app with social accountability I've created Foorr, a minimal to-do list app that focuses on short-term getting-things-done and the option to do this with friends by holding each other accountable and helping each other grow together. Main goal is to fuel your progress whatever needs to be done and hold yourself accountable for your own progress (with some external motivation). Here's what it does: - Create tasks for today and tomorrow only (real hyperfocus on short term GTD). - Invite friends to cheer each other moving forward. - Building up a daily streak as you finish all tasks daily. All tasks completed before midnight which were planned for that day, earns you a level up. Rewarding that sense of completion. Why I built it: I used physical post-it notes and was a bit done with it. I really wanted something minimal focusing only on my to-do's for today and tomorrow. No bloated features, nothing to fancy/polished, just something no-nonsense I wanted to use myself. It might be useful for others who struggle with procrastination and keeping op progress getting things actually done. Feedback is definitely welcome and nice to hear if this resonates with anyone else. https://foorr.com/ July 5, 2024 at 12:14AM
Thursday, July 4, 2024
Wednesday, July 3, 2024
Show HN: Open Sourcing Our No-Code WebXR Editor After 5 Years of Development https://ift.tt/bKs1vPR
Show HN: Open Sourcing Our No-Code WebXR Editor After 5 Years of Development Transfer Thought is a No-Code platform that makes it so anyone can build VR apps directly in their browser. We started this company part-time, building it during commutes to and from work on the train. Over the last 5 years, we've experienced many ups and downs: - Gained early customers - Quit our day jobs - Secured angel funding - Survived with a short runway - Accepted into Techstars Chicago - Survived with a short runway (again) - Landed our biggest client ever, a Fortune 100 company - Despite our highest revenue, our burn rate caught up to us We looked at different ways to wind down the company and ultimately felt open sourcing the platform was the best way to do right by our customers. Now, anyone who is interested in starting a VR company or just building an app can pick up where we left off. I'm excited about this space, if you need help with a VR app, or want to talk tech, please reach out. Check out the repo: https://ift.tt/wT1reH0 Contact me at keenan [at] transferthought [dot] com. https://ift.tt/wT1reH0 July 4, 2024 at 12:33AM
Tuesday, July 2, 2024
Show HN: Free AI productivity framework for solo founders (with GPT-4) https://ift.tt/fwVWTFm
Show HN: Free AI productivity framework for solo founders (with GPT-4) https://ift.tt/QuNn5Pj July 2, 2024 at 10:43PM
Monday, July 1, 2024
Show HN: I built sales API, Stripe Atlas closed me https://ift.tt/7nyOiu3
Show HN: I built sales API, Stripe Atlas closed me My Co-founder and I have spent the last three months developing customsalesapi.com, a platform designed to automate sales and lead generation processes. Essentially, we streamline everything you'd typically do with Apollo by describing your company to our AI. If you're interested, I'd be thrilled to have you as one of our early clients. Please let me know if you'd like more information. Few days ago, we used Stripe Atlas to create our company, and right after company is created, stripe closed our account (account id is acct_1PMyxBIbdJ0fozrG). I went to asking help from many groups but I saw this exact same thing is happening everywhere July 1, 2024 at 11:30PM
Subscribe to:
Posts (Atom)