The best Hacker News stories from Show from the past week
Latest posts:
Show HN: AskHN
Show HN: AskHN
Show HN: Yobulk – Open-source CSV importer powered by GPT3
Show HN: Yobulk – Open-source CSV importer powered by GPT3
Show HN: Phind.com – Generative AI search engine for developers
Hi HN,<p>Today we're launching phind.com, a developer-focused search engine that uses generative AI to browse the web and answer technical questions, complete with code examples and detailed explanations. It's version 1.0 of what was previously known as Hello (beta.sayhello.so) and has been completely reworked to be more accurate and reliable.<p>Because it's connected to the internet, Phind is always up-to-date and has access to docs, issues, and bugs that ChatGPT hasn't seen. Like ChatGPT, you can ask followup questions. Phind is smart enough to perform a new search and join it with the existing conversation context. We're merging the best of ChatGPT with the best of Google.<p>You're probably wondering how it's different from the new Bing. For one, we don't dumb down a user's query the way that the new Bing does. We feed your question into the model exactly as it was asked, and are laser-focused on providing developers the most detailed and comprehensive explanations to code-related questions. Secondly, we've focused the model on providing answers instead of chatbot small talk. This is one of the major improvements we've made since exiting beta.<p>Phind has the creative abilities to generate code, write essays, and even compose some poems/raps but isn't interested in having a conversation for conversation's sake. It should refuse to state its own opinion and rather provide a comprehensive summary of what it found online. When it isn't sure, it's designed to say so. It's not perfect yet, and misinterprets answers ~5% of the time. An example of Phind's adversarial question answering ability is <a href="https://phind.com/search?q=why+is+replacing+NaCL+with+NaCN+in+a+cooking+recipe+good+for+you">https://phind.com/search?q=why+is+replacing+NaCL+with+NaCN+i...</a>.<p>ChatGPT became useful by learning to generate answers it thinks humans will find helpful, via a technique called Reinforcement Learning from Human Feedback (RLHF). In RLHF, a model generates multiple candidate answers for a given question and a human rates which one is better. The comparison data is then fed back into the model through an algorithm such as PPO. To improve answer quality, we're deploying RLAIF — an improvement over RLHF where the AI itself generates comparison data instead of humans. Generative LLMs have already reached the point where they can review the quality of their own answers as good or better than an average human rater tasked with annotating data for RLHF.<p>We still have a long way to go, but Phind is state-of-the-art at answering complex technical questions and writing intricate guides all while citing its sources. We'd love to hear your feedback.<p>Examples:<p><a href="https://phind.com/search?q=How+to+set+up+a+CI/CD+pipeline+in+GitLab+step-by-step">https://phind.com/search?q=How+to+set+up+a+CI/CD+pipeline+...</a><p><a href="https://phind.com/search?q=how+to+debug+pthread+race+conditions+in+c++">https://phind.com/search?q=how+to+debug+pthread+race+conditi...</a><p><a href="https://phind.com/search?q=example+of+a+c+++semaphore">https://phind.com/search?q=example+of+a+c+++semaphore</a><p><a href="https://phind.com/search?q=What+is+the+best+way+to+deploy+a+transformer+for+inference?">https://phind.com/search?q=What+is+the+best+way+to+deploy+a+...</a><p><a href="https://phind.com/search?q=show+me+when+to+use+defaultdicts+over+regular+dicts">https://phind.com/search?q=show+me+when+to+use+defaultdicts+...</a><p>Discord: <a href="https://discord.gg/qHj8pwYCNg" rel="nofollow">https://discord.gg/qHj8pwYCNg</a>
Show HN: Phind.com – Generative AI search engine for developers
Hi HN,<p>Today we're launching phind.com, a developer-focused search engine that uses generative AI to browse the web and answer technical questions, complete with code examples and detailed explanations. It's version 1.0 of what was previously known as Hello (beta.sayhello.so) and has been completely reworked to be more accurate and reliable.<p>Because it's connected to the internet, Phind is always up-to-date and has access to docs, issues, and bugs that ChatGPT hasn't seen. Like ChatGPT, you can ask followup questions. Phind is smart enough to perform a new search and join it with the existing conversation context. We're merging the best of ChatGPT with the best of Google.<p>You're probably wondering how it's different from the new Bing. For one, we don't dumb down a user's query the way that the new Bing does. We feed your question into the model exactly as it was asked, and are laser-focused on providing developers the most detailed and comprehensive explanations to code-related questions. Secondly, we've focused the model on providing answers instead of chatbot small talk. This is one of the major improvements we've made since exiting beta.<p>Phind has the creative abilities to generate code, write essays, and even compose some poems/raps but isn't interested in having a conversation for conversation's sake. It should refuse to state its own opinion and rather provide a comprehensive summary of what it found online. When it isn't sure, it's designed to say so. It's not perfect yet, and misinterprets answers ~5% of the time. An example of Phind's adversarial question answering ability is <a href="https://phind.com/search?q=why+is+replacing+NaCL+with+NaCN+in+a+cooking+recipe+good+for+you">https://phind.com/search?q=why+is+replacing+NaCL+with+NaCN+i...</a>.<p>ChatGPT became useful by learning to generate answers it thinks humans will find helpful, via a technique called Reinforcement Learning from Human Feedback (RLHF). In RLHF, a model generates multiple candidate answers for a given question and a human rates which one is better. The comparison data is then fed back into the model through an algorithm such as PPO. To improve answer quality, we're deploying RLAIF — an improvement over RLHF where the AI itself generates comparison data instead of humans. Generative LLMs have already reached the point where they can review the quality of their own answers as good or better than an average human rater tasked with annotating data for RLHF.<p>We still have a long way to go, but Phind is state-of-the-art at answering complex technical questions and writing intricate guides all while citing its sources. We'd love to hear your feedback.<p>Examples:<p><a href="https://phind.com/search?q=How+to+set+up+a+CI/CD+pipeline+in+GitLab+step-by-step">https://phind.com/search?q=How+to+set+up+a+CI/CD+pipeline+...</a><p><a href="https://phind.com/search?q=how+to+debug+pthread+race+conditions+in+c++">https://phind.com/search?q=how+to+debug+pthread+race+conditi...</a><p><a href="https://phind.com/search?q=example+of+a+c+++semaphore">https://phind.com/search?q=example+of+a+c+++semaphore</a><p><a href="https://phind.com/search?q=What+is+the+best+way+to+deploy+a+transformer+for+inference?">https://phind.com/search?q=What+is+the+best+way+to+deploy+a+...</a><p><a href="https://phind.com/search?q=show+me+when+to+use+defaultdicts+over+regular+dicts">https://phind.com/search?q=show+me+when+to+use+defaultdicts+...</a><p>Discord: <a href="https://discord.gg/qHj8pwYCNg" rel="nofollow">https://discord.gg/qHj8pwYCNg</a>
Show HN: Turn your Pandas dataframe into a Tableau-style UI for visual analysis
Hey, guys. I've just made a plugin which turns your pandas dataframe into a tableau-style component. It allows you to explore the dataframe with easy drag-and-drop UI.<p>You can use PyGWalker in Jupyter, Google Colab, or even Kaggle Notebook to easily explore your data and generate interactive visualizations.<p>PyGWalker (pronounced like "Pig Walker", just for fun) is named as an abbreviation of "Python binding of Graphic Walker".<p>Here are some links to check it out:<p>The Github Repo: <a href="https://github.com/Kanaries/pygwalker">https://github.com/Kanaries/pygwalker</a><p>Use PyGWalker in Kaggle: <a href="https://www.kaggle.com/asmdef/pygwalker-test" rel="nofollow">https://www.kaggle.com/asmdef/pygwalker-test</a><p>Feedback and suggestions are appreciated! Please feel free to try it out and let me know what you think. Thanks for your support!
Show HN: Turn your Pandas dataframe into a Tableau-style UI for visual analysis
Hey, guys. I've just made a plugin which turns your pandas dataframe into a tableau-style component. It allows you to explore the dataframe with easy drag-and-drop UI.<p>You can use PyGWalker in Jupyter, Google Colab, or even Kaggle Notebook to easily explore your data and generate interactive visualizations.<p>PyGWalker (pronounced like "Pig Walker", just for fun) is named as an abbreviation of "Python binding of Graphic Walker".<p>Here are some links to check it out:<p>The Github Repo: <a href="https://github.com/Kanaries/pygwalker">https://github.com/Kanaries/pygwalker</a><p>Use PyGWalker in Kaggle: <a href="https://www.kaggle.com/asmdef/pygwalker-test" rel="nofollow">https://www.kaggle.com/asmdef/pygwalker-test</a><p>Feedback and suggestions are appreciated! Please feel free to try it out and let me know what you think. Thanks for your support!
Show HN: Noya – A new kind of design tool
Hi HN. I've been working on a new browser-based design tool that's ready for you to try.<p>The idea is you work on your design in low fidelity wireframes, while still getting a high fidelity output that you can share or use as a reference for your implementation. The way it works is by mapping low fidelity blocks you draw into high fidelity design system & React components.<p>I spent several years working on design tools at companies like Airbnb, and I think the ideas behind many of the tools we built for designing at scale could really help startups and small teams as well. I would love any feedback you have!<p>PS: Most of Noya is open source at <a href="https://github.com/noya-app/noya">https://github.com/noya-app/noya</a>
Show HN: Noya – A new kind of design tool
Hi HN. I've been working on a new browser-based design tool that's ready for you to try.<p>The idea is you work on your design in low fidelity wireframes, while still getting a high fidelity output that you can share or use as a reference for your implementation. The way it works is by mapping low fidelity blocks you draw into high fidelity design system & React components.<p>I spent several years working on design tools at companies like Airbnb, and I think the ideas behind many of the tools we built for designing at scale could really help startups and small teams as well. I would love any feedback you have!<p>PS: Most of Noya is open source at <a href="https://github.com/noya-app/noya">https://github.com/noya-app/noya</a>
Epstein Documents, Fully Searchable
Show HN: I made an early 2000s-inspired internet forum
Show HN: I made an early 2000s-inspired internet forum
Show HN: Ractor – a Rust-based actor framework with clusters and supervisors
SOME HIGHLIGHTS.<p>* We have a full supervision tree so actors can monitor other actors for exits and unhandled panic!s (at least the ones that can be caught)<p>* The actor lifecycle is handled for you in a simple single-threaded, message handler primitive<p>* You have a mutable state with each message handling call, so you have an easy way to create stateful actors and update that state as messages are processed<p>* Actors talk to other actors by message passing, but there are remote-procedure-calls (RPCs) so actors can "ask a question" to another actor and wait on the reply.<p>* A lot of the concurrency primitives are handled by the framework, such as cancellation/termination of actors (both graceful and forceful)<p>* A Factory primitive in order to formulate distributed processing pools with multiple job routing options<p>* Early but stable support for a distributed epmd-like cluster environment, where you can talk to actors over a network link. It's an additional crate (ractor_cluster) that builds on ractor to facilitate the inter-connection between nodes and support remote casts and calls to actors on a remote node.<p>We're openly seeking feedback, so please feel free to utilize the library and let us know if there's anything you find missing or doesn't work as expected!
Show HN: I built a little online drum machine using 808 style samples
Show HN: Self-host Whisper As a Service with GUI and queueing
Schibsted created a transcription service for our journalists to transcribe audio interviews and podcasts really quick.
Show HN: I made Hacker News but for research papers
Show HN: DSLCad – a programming language and interpreter for building 3D models
Hey HN.
Over the last half a year I have been working on a 3D CAD programming language called DSLCAD. Today I am here to show my first release!<p>It is heavily inspired by OpenSCAD which got me hooked on the idea of a CAD programming language and what it can do.<p>Please let me know what you think. Ill be in the thread to answer any questions you may have.
Show HN: Spaghettify – A VSCode Extension to make your code worse with AI
Recently, I saw Github Next Code Brushes (<a href="https://githubnext.com/projects/code-brushes" rel="nofollow">https://githubnext.com/projects/code-brushes</a>) which is a VSCode extension that can use AI to do things like make your code more readable, add types, etc.<p>This inspired me: what if there was an extension that could use AI to make your code worse as well?<p>Meet Spaghettify. An extension I developed that can do things like:<p>- Obscure your code<p>- Introduce a bug<p>- Make the variable names way too descriptive<p>- Use nonsymmetrical whitespace<p>- Add irrelevant/rambling comments<p>- Document the code with emoji<p>- Document the code in any style (Dirty Limerick, Fast Talkin' 1930s Gangster)<p>I built this extension over the last couple of days and released it in collaboration with BCAD.one. It's available on the VSCode Marketplace, you will need an OpenAI API Key to use it. You can grab it here:<p>(<a href="https://marketplace.visualstudio.com/items?itemName=BCAD.spaghettify" rel="nofollow">https://marketplace.visualstudio.com/items?itemName=BCAD.spa...</a>)<p>Code here:<p><a href="https://github.com/BeforeCutieAfterDoggo/spaghettify">https://github.com/BeforeCutieAfterDoggo/spaghettify</a>
Show HN: boxxy – Control where Linux programs put files, without symlinks