Monday, September 30, 2024
Sunday, September 29, 2024
Show HN: I Built a Multimodel AI Chatbot to Avoid Having FOMO on the Best Models https://ift.tt/6pDKUOt
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 08:08PM
Saturday, September 28, 2024
Show HN: Modern Benchmarking Tooling for JavaScript https://ift.tt/unp6JhY
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/Oap3Znd September 28, 2024 at 10:54PM
Friday, September 27, 2024
Thursday, September 26, 2024
Wednesday, September 25, 2024
Show HN: Httpdbg – A tool to trace the HTTP requests sent by your Python code https://ift.tt/MjStdCu
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/c7JtVBi Documentation: https://ift.tt/Hjroh4s A blog post on a use case: https://ift.tt/bZgxfh3... https://ift.tt/c7JtVBi September 25, 2024 at 11:18PM
Tuesday, September 24, 2024
Show HN: Agentic AI coding tool that's easy to use, transparent and customizable https://ift.tt/RZOjvrQ
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/LwBne2K 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/2RnsvPK . 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/Oxb0JKN https://ift.tt/NbOpjst https://ift.tt/FWeuNrG September 24, 2024 at 11:47PM
Monday, September 23, 2024
Show HN: I made a small platform to collect feedback https://ift.tt/hVs36Y7
Show HN: I made a small platform to collect feedback Hello everyone! My co-founder and I are developing a mid-sized travel platform. (By the way, we met on cofounder-matching, an excellent platform) In the process of development, we decided to create a feature to collect customer reviews on the site. It's simple enough: You put a rating in stars and write a review. It seemed incredible to me to launch this project separately from the main project because, in my experience of working in companies, it was often done again. I started this project as a test run for Indie Making. Right now, the functionality is pretty modest (as well as the cost); you put in a script, collect data, and analyze it. If you have any ideas on how to improve the project, what to pay attention to, or what functionality is sorely lacking, I will be glad to listen and discuss them. Thank you! https://ift.tt/XSIBYfU September 23, 2024 at 11:30PM
Show HN: DataToaster 3000 – The Future of Kitchen Data Storage https://ift.tt/TlkxSAs
Show HN: DataToaster 3000 – The Future of Kitchen Data Storage https://ift.tt/bueJ28V September 23, 2024 at 11:05PM
Sunday, September 22, 2024
Show HN: LyricFluent – Learn Languages with Music (Mainly Spanish) https://ift.tt/sNGLy1h
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/yJecsC5 September 22, 2024 at 05:23PM
Show HN: I built a tool to stop chasing users for feedback and screenshots https://ift.tt/qJDwj0K
Show HN: I built a tool to stop chasing users for feedback and screenshots Hey HN, I got fed up with chasing users for screenshots and building custom feedback forms for every app, so I built FeedbackThing.Pro to make it easier. Now, all I do is drop in a simple
Saturday, September 21, 2024
Friday, September 20, 2024
Show HN: EloqKV – Scalable distributed ACID key-value database with Redis API https://ift.tt/dRZlrfa
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/G192TV6 September 19, 2024 at 04:03PM
Thursday, September 19, 2024
Show HN: Keep Your Next Viral AI App Free for Longer with Local Embeddings https://ift.tt/v89lyod
Show HN: Keep Your Next Viral AI App Free for Longer with Local Embeddings Hey HN! I'm the founder of Function, a platform that enables developers to run Python functions on-device. We've been quietly building for the past several months, but we figured we could launch some low-hanging fruit: Function LLM patches your OpenAI client to generate embeddings on-device, both in the browser and in Node.js. The library itself is so tiny that it doesn't even need to be a standalone library. The more interesting piece is how it uses Function to generate embeddings on-device, fully cross-platform. For the embedding model, we've partnered with Nomic AI to use their `nomic-embed-text-v1.5` model. We plan to add more embedding models before adding support for text generation with small LLMs. At this point, you're probably wondering how we differ from the likes of Ollama. Unlike the other players in the space, we believe that the whole point of on-device AI is to push as much of your app's heavy computation to your users' devices--not to spin up a 'local' AI service still running on AWS or other cloud. This way, you save tons of cash by simply not having a hosted inference service; build users' trust with privacy by default; and have way less complicated serving infrastructure at scale. Function handles much of the heavy lifting for you. We compile Python functions to run on Android, browser, iOS, macOS, Linux, and Windows. And we're exhaustive in using whatever hardware (GPU, NPU, etc) or ISA (CoreML, CUDA, etc) a particular device offers. --- Relevant Links: - Function LLM on GitHub (please ): https://ift.tt/tfNClQp - Document Retrieval Demo (fully on-device): https://ift.tt/1F3kc4H - Function Docs: https://ift.tt/jCtWdDO - Function Waitlist (to write your own fxns): https://fxn.ai/waitlist https://ift.tt/gD93xEV September 19, 2024 at 11:51PM
Show HN: An Elliptic Curve-Based Secure Chat, Written Using Rust and Protobuf https://ift.tt/4qjrFp9
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/5qiv7UR ) 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/zUAIgq3 September 19, 2024 at 10:22PM
Wednesday, September 18, 2024
Tuesday, September 17, 2024
Show HN: shadcn-chat-cli – customizable chat react components https://ift.tt/2IrOhVW
Show HN: shadcn-chat-cli – customizable chat react components https://ift.tt/OxDsILz September 17, 2024 at 10:52PM
Monday, September 16, 2024
Sunday, September 15, 2024
Show HN: Subdomain Enumeration CLI https://ift.tt/CY5IyxR
Show HN: Subdomain Enumeration CLI https://ift.tt/MUuN9yb September 15, 2024 at 10:12PM
Saturday, September 14, 2024
Show HN: Preppr – a modern prepping/survivalist iOS app https://ift.tt/EZf1WOH
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/haq6fd5 September 14, 2024 at 10:13PM
Friday, September 13, 2024
Thursday, September 12, 2024
Wednesday, September 11, 2024
Show HN: Mandoline – Custom LLM Evaluations for Real-World Use Cases https://ift.tt/XRvzHGP
Show HN: Mandoline – Custom LLM Evaluations for Real-World Use Cases Hi HN! We're a small team of AI engineers who've spent the last few years building AI applications. Through this, we've experienced firsthand many of the challenges that come with evaluating and improving AI systems in real-world contexts. Standard LLM evaluations (and evaluation methods) often use simplified scenarios that don't reflect the complexity LLMs encounter in actual use. This leads to a disconnect between reported performance and real-world usefulness. We built Mandoline to bridge this gap, helping developers evaluate and improve LLM applications in ways that matter to end-users. Our approach allows you to design custom evaluation criteria that align with your specific product requirements. For a quick overview of how it works, check out our Python and / or Node SDK READMEs: - Python: https://ift.tt/GUeTpWZ... - Node / Typescript: https://ift.tt/6VKAtZa... Hopefully this design is flexible yet scalable, and helps you do things like: track LLM progress over time, make informed AI system design decisions, choose the best model for your use case, prompt engineer more systematically, and so on. Under the hood, Mandoline is a hybrid system using a combination of our own models and top general-purpose LLM APIs. We used Mandoline to evaluate and improve itself, which helped us make better decisions about system design. In the future, we’ll be adding visualization tools to more easily analyze trends, and expanding our in-house models capabilities to reduce reliance on (and hopefully outperform) external models. Check out our website ( https://mandoline.ai/ ) and documentation ( https://ift.tt/FhNO8m0 ) to learn more. We’d love to hear about your experiences with evaluating AI systems for production use. What have you found most challenging in evaluating AI systems? What behaviors are hard to quantify? How could Mandoline fit into your workflow? You can reach us here in the comments or send us an email (hi@mandoline.ai). We appreciate you taking the time to learn a bit about Mandoline. https://mandoline.ai September 11, 2024 at 11:03PM
Tuesday, September 10, 2024
Monday, September 9, 2024
Sunday, September 8, 2024
Saturday, September 7, 2024
Friday, September 6, 2024
Thursday, September 5, 2024
Wednesday, September 4, 2024
Tuesday, September 3, 2024
Monday, September 2, 2024
Show HN: Full Text, Full Archive RSS Feeds for Any Blog https://ift.tt/zyqn8Xw
Show HN: Full Text, Full Archive RSS Feeds for Any Blog https://ift.tt/YAf57xG September 2, 2024 at 05:06PM
Subscribe to:
Posts (Atom)