Show HN: Create books on any topic with ChatGPT https://bit.ly/3Ku3vMg

Show HN: Create books on any topic with ChatGPT This weekend wanted to try out the streaming ChatGPT API. As test project, created a simple website, which allows you to create whole books by providing just the title (with optional extra instructions). Given the title of the book, table of contents, chapters and sections with text content are generated. All books are listed on the home page, sorted into topic categories. As a bonus, when reading a specific section of a chapter, you can use the chat interface to ask for clarifications, request more content, list relevant topics, etc. The system does quite well with knowledge based books (history for example) and not so good with literature, where more prompt tweaking is needed. What books would you create? # Examples: The Bible https://bit.ly/3KrrY4M Child Psychology for Parents https://bit.ly/3KujSIx ChatGPT: A step towards Artificial General Intelligence https://bit.ly/3DMxdrY The World History from a global point of view https://bit.ly/3DLzMdZ The History of Ukraine https://bit.ly/3Ku3yYs Art of War: A modern perspective https://bit.ly/3DL9YP2 Kamasutra: A modern perspective https://bit.ly/3DLrrXz https://bit.ly/3Ku3Bn6 August 6, 2023 at 04:07PM

Show HN: A guide to self-host AudioCraft demo https://bit.ly/3DLP4Q4

Show HN: A guide to self-host AudioCraft demo https://bit.ly/47f0nO9 August 6, 2023 at 05:41AM

Show HN: Archsense – Accurately generated architecture from the source code https://bit.ly/3OLyDti

Show HN: Archsense – Accurately generated architecture from the source code https://bit.ly/47yGZMj August 6, 2023 at 10:47AM

Show HN: Postgres Language Server https://bit.ly/3ql4mbe

Show HN: Postgres Language Server https://bit.ly/47hVhAx August 6, 2023 at 11:26AM

Show HN: Briefed – Summaries for Hard Paywalled Content https://bit.ly/45bnDeB

Show HN: Briefed – Summaries for Hard Paywalled Content Hey HN! Briefed developer here. Briefed creates summaries for (mostly hard) paywalled content. I built it because it's something that I wanted, but couldn't find (specifically the hard paywalled part). Why did I want this? 1. There are a lot of publications that I'm only tangentially interested in. Not enough to warrant a full subscription but enough to want to get the gist of what they're saying. A step between reading the title and reading the full article. Here's a list of the publications we currently create summaries for: https://bit.ly/44W9nGk 2. I often don't have the time to read entire articles. I can save an article as "read later" in my rss app, but I've often lost interest by the time I get to it. Briefed creates two summaries for each article: one detailed, and one tldr. Here is an example of one: https://bit.ly/47huEf3... 3. Sometimes I've read the full paywalled article and want to read comments here on HN, but hard paywalls limit the engagement those posts will get. Hopefully a detailed summary is enough to start meaningful and curious conversation. I'm still kicking the tires, so the first 500 people to sign up will get the first year free (normal price is $12/year; the price that I felt I would personally pay for such a service). Use the promo code LAUNCH500 at checkout. The code will be automatically invalidated after 500 people use it. Let me know what you think! Any feedback , issues, or questions are welcome. Thanks, Brandon https://bit.ly/3qlBmjE August 5, 2023 at 08:23PM

Show HN: Shapetoy – an application to combine shapes or create shape animations https://bit.ly/3OKutBJ

Show HN: Shapetoy – an application to combine shapes or create shape animations https://bit.ly/3qj7Pam August 5, 2023 at 05:29PM

Show HN: Embedding hidden text in AI images with ControlNet and Modal https://bit.ly/47nuF12

Show HN: Embedding hidden text in AI images with ControlNet and Modal A few weeks ago, a technique was shared for hiding text in the negative space of an AI-generated image and I thought it was an amazing illusion. However, the only tutorial I could find was light on details and UI-based (and seems to have disappeared?). To continuing exploring on my own, I wrote a script to generate this style of images programmatically using various Stable Diffusion checkpoints and a QR code-optimized ControlNet, running on Modal. Let me know if it works for you! https://bit.ly/3OIxHET August 5, 2023 at 04:45PM

Show HN: Unstock.ai – Free image generation with SDXL https://bit.ly/3DIhWZd

Show HN: Unstock.ai – Free image generation with SDXL Hi HN! I've got the barebones of a service running on top of Stable Diffusion XL. I can cheaply run image generations at 1024x1024. And of course there's a limit to how fast I can generate them given the request queue and limited GPUs, but the service is cheap enough that I'm happy to run it out of pocket for now. Let me know your thoughts, I hope you enjoy the service! https://bit.ly/3QrZBY8 August 5, 2023 at 03:02AM

Show HN: Matrix. A Sci-Fi Comic https://bit.ly/3Yp0C58

Show HN: Matrix. A Sci-Fi Comic On the occasion of completing 100 short stories in Matrix, I thought of sharing it with the wider audience. Matrix is a culmination of my 9 months of travel across India and 2 days that I spent in Barcelona. I started the project with an aim to improve my writing. But now I feel like the project is directing me on how it wants to take shape. And as I continued, The project took a shape of its own. I believe it is in a good enough shape to be considered as a mockumentary on life and meaning. It still only has me as a primary contributor because of which it has my limitations and my biases. If you like to contribute, Please feel free to create a pull request on the दुनिया(World in Hindi) branch. जय श्री राम| https://bit.ly/3Qsbtcx August 5, 2023 at 10:57AM

Show HN: Custom Haskell handlers for Nginx https://bit.ly/3YoTpSC

Show HN: Custom Haskell handlers for Nginx This is rather a mature project. It began out of curiosity: I wanted to test if Haskell FFI was powerful and expressive enough to interconnect C and Haskell code flawlessly. Particularly, if C strings generated inside Nginx can be shared within Haskell code and what must be done to respect their lifetimes etc. Recently, I released version 3.2.0 with revamped README (with a lot of examples) and a new approach to building Haskell handlers using the modernized cabal v2-build for dependencies. https://bit.ly/43UdukY August 4, 2023 at 12:19PM

Show HN: Open-access book on platform governance https://bit.ly/3OlVLx0

Show HN: Open-access book on platform governance Hi there, I'm a political theorist, a professor at Northwestern law school and a founding fellow, board member, etc. of the Integrity Institute. I also was the in-house democratic theorist on Facebook's "civic integrity" (election protection) team for a while, and helped out with the research supporting the creation of the Meta Oversight Board. I just published an academic book with Cambridge University Press bringing together those experiences with research in political science to argue that the way forward for mitigating the harms of big internet platforms and governing user behavior on them is to create global direct democratic institutions for them. You can read the whole book for free on Cambridge's website or by downloading the open-access (license CC-BY-NC-ND 4.0) PDF at the link. If you want a hard copy, you can also buy it in paperback or hardcover. https://bit.ly/3qqTCrL August 4, 2023 at 07:51PM

Show HN: Open-source library to manage todo items in Python code https://bit.ly/3Ym91Gs

Show HN: Open-source library to manage todo items in Python code I made this library that collects all TODO items from a python project, creates automated summaries and notifies them over email, SNS or SES etc. It is highly customizable allowing to add custom summaries, custom notifier and integration with downstream applications. https://bit.ly/3KqcFsY August 4, 2023 at 06:34PM

Show HN: YC idea matcher – Submit an idea and get a list of similar YC companies https://bit.ly/3OHVpC1

Show HN: YC idea matcher – Submit an idea and get a list of similar YC companies This project uses semantic search, an advanced search technique that aims to understand the intent and context behind a search query instead of just matching keywords. It's built using Neon Postgres + pg_embedding and OpenAI for generating embeddings. More details can be found in the repo https://bit.ly/440prWa https://bit.ly/3qgSJSz August 4, 2023 at 05:15PM

Show HN: SymbolicAI https://bit.ly/43U9mRU

Show HN: SymbolicAI The SymbolicAI project started somewhere at the end of the last year and had its first commit mid January this year. If I would to be briefly summarize "why" do we think it's a project worth working on, is because of the following idea: we're slowly marching towards software 3.0 and we need to grow frameworks to a maturity point that would allow people not only to PoC their own ideas, but also to gain access to a strong community support that nurtures the mutual exchange of ideas between the individuals. I personally believe this is the secret behind many successful OS projects (e.g. Neovim, LazyGit, PyTorch, Jax, just to name a few). FAQ Q: What the project does? A: A lot. You can build your own chatbot, interact with as many as 13 tools (search, wolfram, dall-e, blip, clip, ocr, pinecone, whisper, selenium, local files, etc.), pretty much most of the things you've already seen hyped or cool on the social media. Q: Sounds close to… LangChain…? A: Briefly, I think LangChain grew too fast and became the jack of all trades but the master of none. I'm sure they had their reasons for approaching things the way they did, and I don't want to make this post about them more than I already have. Others have had more thorough investigations of this topic and better rants than I would. Q: Ok, then tell me why would I want to be part of it? A: We're x2 core developers. Sometimes less is more, giving us time to think more deeply about designing the framework and making it accessible to others. Some principles: - Ease of use and flexibility: we were heavily inspired by PyTorch, and we aimed to follow the same code structure one uses with torch. Our original intuition was that when you're introducing something new, tying it with something people are familiar with will make it more accessible (in terms of read/write). Not only this, but the initial recipe proved quite successful and replacing it with something else without concrete reasons is not worth doing IMHO. Moreover, one of our long-term visions is to have smooth integration with torch. We aim to grant SymbolicAI differentiable features. Imagine your chatbot learning to better use its memory (e.g. how to update its memory with relevant information). - As in torch everything is a tensor, in our framework everything is a Symbol. A Symbol once defined gets accessed to some primitives (as an analogy think of PrimTorch) which would easily allow you to compose complex expressions or manipulate Symbol variables. This unlock very fast manipulations (i.e. dot notation <|object|><|dot|><|method|>). - The hard work is done by decorators. We use them for the following reasons: (1) modularity, (2) composition, (3) flexibility, and (4) readability. - We want to make a cohesive dev environment. I'm a script kiddo and I don't like to leave my terminal. I dislike web interfaces. I want to use my local env with my own setup. We have an experimental feature that is built on top of git and would enable package management. It's similar to pip, but for extensions built with our framework. Another long-term vision is to make accessible to anyone using our framework a quick share with the community. See https://bit.ly/3YBq9Z7 for a showcase of how to do transcription and create youtube chapters with Whisper with our package manager. There's much more to say, but I will stop here. Please check our GitHub README ( https://bit.ly/43Q6rtz ) for a more deep dive or our latest tutorial video that highlights some relevant use-cases from a more high-level POV ( https://www.youtube.com/watch?v=0AqB6SEvRqo ). I really do hope that at least some of you reading will get interested. We have so many goals we want to reach, so many ideas we want to test, and probably just as many bugs (we call them maggots just for fun) we need to fix. We need you. https://bit.ly/43Q6rtz August 4, 2023 at 09:51AM

Show HN: Zep – pgvector-based memory store for LLM apps https://bit.ly/3Ok4JuK

Show HN: Zep – pgvector-based memory store for LLM apps Hey HN - we launched Zep's document vector DB today. Zep is an open source memory store for LLM apps, and this builds on existing chat history memory persistence, embedding, and enrichment capabilities. Zep uses Postgres and pgvector for database operations and vector search. Vector search can be complicated on Postgres, with careful configuration required at both index creation and query time. We've focused on significantly improving this developer experience. Zep automatically selects index and query parameters for developers based on best practices and known heuristics. Vector database operations are exposed via a simple Python (and LangChain) API for working with document collections, documents, and search results. While we focus on LLM App use cases, you can turn any Postgres instance with pgvector into a vector database with great DX. Our launch announcement: https://bit.ly/3Kpfr1P... -Daniel https://bit.ly/3DETQib August 3, 2023 at 07:17PM

Show HN: Create your first ZK-SNARK Contract with Mina Blockchain https://bit.ly/3Yn2mMn

Show HN: Create your first ZK-SNARK Contract with Mina Blockchain https://bit.ly/3rTUe9Y August 3, 2023 at 07:21PM

Show HN: I made Grammarly for accessibility code violations https://bit.ly/3OHvbzF

Show HN: I made Grammarly for accessibility code violations https://bit.ly/3qgVGCJ August 3, 2023 at 06:37PM

Show HN: WebAssembly dev environment for Envoy Proxy https://bit.ly/47b87Rf

Show HN: WebAssembly dev environment for Envoy Proxy Hi HN! For the past few weeks we've been working on Proximal - a workflow engine that lets you quickly iterate on WebAssembly extensions for Envoy Proxy[0] (or other proxies) right on your local machine: https://bit.ly/3KpOaw6 This work is based on Proxy-WASM[1] extension ABI for Envoy (and other proxies like APISIX and Mosn[2]) which allows you to execute WebAssembly code on every API request a la Cloudflare Workers. As part of our wider effort at https://bit.ly/3YgVZtX to improve API glue code we built an experimentation / development platform and hope you will find it useful! On the technical side this project packs Envoy itself, Envoy controller, REST API (for controlling the controller =)), React SPA, and Temporal server/worker (for orchestration) - all baked into a single Go binary. You can find more on architecture and limitations in the repository README[4]. This project is pretty early stage and we would appreciate community feedback! Previous HN discussions on this topic: * https://bit.ly/3QGZcBl * https://bit.ly/3Kqjlr3 --- [0] https://bit.ly/3rJ29qp [1] https://bit.ly/3YmaBrP... [2] https://bit.ly/3QpsWSR https://bit.ly/45awMU6 [3] https://bit.ly/3QlknbR... https://bit.ly/3KpOaw6 August 3, 2023 at 05:39PM

Show HN: Hydra 1.0 – open-source column-oriented Postgres https://bit.ly/3YiNU82

Show HN: Hydra 1.0 – open-source column-oriented Postgres hi hn, hydra ceo here hydra is an open source, column-oriented postgres. you can set up remarkably fast aggregates on your project in minutes to query billions of rows instantly. postgres is great, but aggregates can take minutes to hours to return results on large data sets. long-running analytical queries hog database resources and degrade performance. use hydra to run much faster analytics on postgres without making code changes. data is automatically loaded into columnar format and compressed. connect to hydra with your preferred postgres client (psql, dbeaver, etc). following 4 months of development on hydra v0.3.0-alpha, our team is proud to share our first major version release. hydra 1.0 is under active development, but ready for use and feedback. we’re aiming to release 1.0 into general availability (ga) soon. for testing, try the hydra free tier to create a column-oriented postgres instance on the cloud. https://bit.ly/3QlkjJ9 https://bit.ly/44TIUZR August 3, 2023 at 05:19PM

Show HN: Making Don Quijote accessible to Spanish learners https://bit.ly/44UsmRs

Show HN: Making Don Quijote accessible to Spanish learners I love simple english wikipedia and I wanted to use that style of writing to make spanish books accessible to me. First I simplified the spanish[0] but simplification can't remove every word I don't know. This led me to adding the definition for words I probably wouldn't know[1]. Second simplifying the spanish ruined the magic of Don Quijote so I tried just adding definitions to the original[2]. The definitions aren't perfect but they let me stay in the flow of the text. I'm tempted to make a firefox extension to annotate any website so I can spend more time reading spanish. Let me know what you think! [0] https://bit.ly/45bjaYS [1] https://bit.ly/3YhDa9T [2] https://bit.ly/457KqqU August 3, 2023 at 04:28PM