prompt
stringlengths 16
15.6k
| completion
stringlengths 4
6
|
|---|---|
Title: Hello HN,<p>My cofounder (jon918) and I started Sym three years ago because we were frustrated with how hard it was to manage access to cloud infrastructure. We wanted to build a tool for JIT access that was <i>actually</i> designed for developers. We were wary of tools that tried to accommodate both devs and IT but ended up with usability compromises for both.<p>First, we figured no one wants another web app to log into so we let administrators define access workflows in Terraform and let developers request and gain access via Slack. That seemed to pay off: being code-based was a big plus for our early customers since it let them manage the logic in version control and test in CI/CD.<p>Second, we knew that updating permissions/roles/access was a major source of toil and risk in the world of cloud infrastructure. Have you ever tried to avoid annoying, persistent access requests by setting policies that are a bit more permissive than you’d like? We felt that fully automated just-in-time access + approvals could really help here. But we also knew that a simple approval tool could end up leading to request fatigue - kind of defeating the purpose. So we built an SDK to let you define checks in code (e.g. pagerduty.on_call, okta.is_user_in_group, github.get_repo_collaborators) in order to dynamically route requests or fast-track access when appropriate. This seems to be paying off: users are creating Slack-based approvals in front of different types of risky actions like production access, sensitive queries and triggering Lambdas.<p>We’d love your feedback on our approach so far. Does this make sense to you? Is this a tool you'd use? What would you want to see out of it?<p>To learn more, check out the video that Nick (nmeans (Sym VPEng)) made [1]. You can also check out our docs [2] or set up your own flow [3].<p>thanks!<p>-adam<p>[1] <a href="https://vimeo.com/815222490/c717c18c42" rel="nofollow">https://vimeo.com/815222490/c717c18c42</a><p>[2] <a href="https://docs.symops.com" rel="nofollow">https://docs.symops.com</a><p>[3] <a href="https://symops.com/signup" rel="nofollow">https://symops.com/signup</a>
Upvote: | 50 |
Title: I remember back in 2017 or 2016 there were posts left and right someone doing an ICO with some breakthrough technology. How have they dissapeared so fast? And why is nobody doing ICOs right now? Were any of the startups that did that successful?
Upvote: | 41 |
Title: I would like to introduce Tabby, which is a self-hosted alternative to GitHub Copilot that you can integrate into your hardware. While GitHub Copilot has made coding more efficient and less time-consuming by assisting developers with suggestions and completing code, it raises concerns around privacy and security.<p>Tabby is in its early stages, and we are excited to receive feedback from the community.<p>Its Github repository is located here: <a href="https://github.com/TabbyML/tabby">https://github.com/TabbyML/tabby</a>.<p>We have also deployed the latest docker image to Huggingface for a live demo: <a href="https://huggingface.co/spaces/TabbyML/tabby" rel="nofollow">https://huggingface.co/spaces/TabbyML/tabby</a>.<p>Tabby is built on top of the popular Hugging Face Transformers / Triton FasterTransformer backend and is designed to be self-hosted, providing you with complete control over your data and privacy. In Tabby's next feature iteration, you can fine-tune the model to meet your project requirements.
Upvote: | 627 |
Title: Machato is a tiny ChatGPT client that supports all basic features, and more !<p>I was frustrated with other native implementations that focused on quick access but didn't give easy access to a history of what has been said. Since ChatGPT is a learning tool for me, I'm always ruffling through past conversations.
The client supports markdown rendering as well as LaTeX. Feel free to try it out !<p>For those who see this thread early, you can use the promo code EARLY_BIRD to get a free lifetime license.<p>Let me know if some aspects can be improved or if there are features you'd like to see implemented in native clients.<p>EDIT: Since the early bird discount has expired before this post could reach HN, I'll leave you with 50% off with the code MACHATO50 !
Upvote: | 68 |
Title: Interested in a HN-like source of information and discussions on AI news. Ideally it would include slightly more in-depth and in the weeds discussions on AI research and developments, while staying away from basic news stories and applications.
Upvote: | 116 |
Title: If you already know what bytebeat is, you don't need an explanation. If not, check my project :)<p>Here is how it looks:<p><pre><code> SELECT mono(output(
arraySum(x -> 1 / 6
* running_envelope(30 * (1 + x / 6), time, 0.05 * x, 0.005, lfo(0, 0.25, sine_wave, time / 8), 0.1)
* sine_wave(time * 80 * exp2(x / 3)),
range(12))))
FROM table;
</code></pre>
To check how it sounds, find the examples in the repository <a href="https://github.com/ClickHouse/NoiSQL">https://github.com/ClickHouse/NoiSQL</a>
Upvote: | 176 |
Title: ChatGDB is a tool designed to superpower your debugging experience with GDB, a debugger for compiled languages. Use it to accelerate your debugging workflow by leveraging the power of ChatGPT to assist you while using GDB!<p>It allows you to explain in natural language what you want to do, and then automatically execute the relevant command. Optionally, you can ask ChatGPT to explain the command it just ran or even pass in any question for it to answer. Focus on what's important - figuring out that nasty bug instead of chasing down GDB commands at the tip of your tongue.<p>See it here: <a href="https://github.com/pgosar/ChatGDB">https://github.com/pgosar/ChatGDB</a>
Upvote: | 189 |
Title: Hi,<p>I'm expecting an offer from an early stage start-up. Would I sound unreasonable to see their cap table? Are there other questions that I can ask that can help me determine what % of the company I'd be getting in this stage?<p>Thank you!
Upvote: | 104 |
Title: I started <a href="https://signalprocessingjobs.com/" rel="nofollow">https://signalprocessingjobs.com/</a> to organize a couple of side projects I have:<p>- a job board for signal processing engineers
- Journal2MATLAB: a blog that turns academic papers into MATLAB code
- some work on using cocotb for verifying DSP HDL
- and some MATLAB code snippets i keep coming back to<p>Hope others in the community find it useful :)
Upvote: | 78 |
Title: Yes, this includes the MBR and partition boot sector. The partition table is modeled as a class hierarchy. Second stage loads a kernel from a FAT32 file.<p>I'm normally a C programmer, but for a work project, I recently revisited C++, a language that I had an on-again-off-again relationship over the years.<p>I did this as a fun training exercise on the side to get familiar with the language again, but obsessed a bit and kind of overshot the target. I thought the result might be of interest to someone.
Upvote: | 64 |
Title: Hey HN!<p>I recently built Prompt Reducer, an app that makes it easier to compress GPT-4 prompts. The main goal is to reduce the number of tokens in each prompt, thereby reducing the cost of running GPT-4. I figured since @gfodor tweeted about compressing GPT-4. It’s still early, and it does not work perfectly, but I’d love to hear any feedback or suggestions for how to make it faster or more efficient.
Upvote: | 68 |
Title: My son had $51K worth of crypto with FTX. On the 9th November he instructed them to sell the coins and obviously received no payment. He just received an email from Kroll Restructuring Administrators showing an amount to his credit of something like $3.00.<p>Are there any other HN readers in similar positions? How can one submit a claim under these circumstances?<p>He is absolutely devastated as this was his only savings and he has been unemployed for the last 12 months. I would appreciate any help and suggestions.
Upvote: | 40 |
Title: Context: I was just listening to the latest Soft Skills Engineering[1] (a podcast I highly recommend, by the way!) and they were consoling a junior eng who bombed an interview by saying everyone has a story about doing this at some point. I thought it'd be nice to crowd-source these stories in case others are feeling bad about their interview experiences :)<p>I can start: My first tech interview ever was the summer going into my Junior year of college. I was still very raw and had no idea what to expect, and got asked a pretty standard string manipulation question. I fumbled around for a while before frantically trying to Google the question in a manner that I now realize was likely very obvious. Needless to say, I didn't get the job. I did learn from the experience, though, and studied my butt off for the next summer, which paid off in much harder interviews :)<p>Looking forward to hearing yours, thanks!<p>[1] <a href="https://softskills.audio/2023/04/03/episode-350-bombing-a-technical-interview-and-background-vetting/" rel="nofollow">https://softskills.audio/2023/04/03/episode-350-bombing-a-te...</a>
Upvote: | 63 |
Title: Hi HN! We just shipped a full-featured TUI (Text User Interface) for chatting with your Marvin bots, powered by GPT 4 or GPT-3.5. Like all of Marvin, it's fully open-source and we hope you find it useful. To launch it, upgrade and run `marvin chat`.<p>The TUI is built with Textual (<a href="https://github.com/textualize/textual/">https://github.com/textualize/textual/</a>) and uses some of its newest features including background workers and modals. We've made base TUIs before but this is the first one that's a true "app" with many screens and coordinated global state. Happy to answer any questions about working with Textual - once it "clicked" it was surprisingly similar to building a traditional front end! Small note: Terminal.app on MacOS isn't great for TUIs, so while it'll work, we suggest an alternative terminal.<p>One of our goals with the TUI was to integrate Marvin's bots into the familiar chat UX. Bots can have distinct personalities, instructions, and use plugins, so each one is like a mini "conversational application." You might know about Marvin because of AI Functions, but at its core Marvin is a library for building and deploying bots (in fact, AI functions are actually a bot!). We started building the TUI as a way to quickly explore and assess our bots' capabilities. It quickly became so useful that we decided to make it a first-class experience.<p>We've preloaded several bots, including one that can guide you through an RPG and another that is obsessed with explaining regex, and will add many more. You can even create your own bots just by asking the default bot (Marvin) to help you.<p>We hope the TUI is a fun way to quickly interact with your bots and it was a great way for us to learn Textual. Please check out the code and let us know what enhancements we can add!
Upvote: | 126 |
Title: hey hn, supabase ceo here<p>this is a postgres connection pooler. it’s similar to pgbouncer, but built with Elixir and specifically designed for multi-tenancy.<p>it’s still under development, but it’s at a stage where we can gather a feedback from the community and you can try it yourself. we aren’t using this in production yet, but aiming to deploy it for a subset of databases in the next 2 months.<p>We have the following benchmarks (details in the readme):<p><pre><code> - Elixir Cluster maintaining 400 connections to a single Postgres database
- 1_000_000 clients connecting to the Elixir cluster
- Sending 20_000 transactions per second
- Consuming 7.8G RAM and ~50% CPU on a 64vCPU machine
</code></pre>
supavisor can be run as a cluster or a single node/binary. It’s handling 90%+ of the throughput of pgbouncer on a local machine (running pgbench)<p>we will place this in front of all supabase databases. It will eventually be able to handle multiple types of connections: traditional TCP connections, and HTTP connections for developers who are connecting to Postgres in serverless environments using Prisma, Kysely, Drizzle, etc<p>the proxy will serve as a connection buffer while we scale databases: scaling up compute with zero-downtime, and for scale-to-zero - triggering a server restart when a connection is initiated<p>finally, i want to shout out to Jose and the Dashbit/elixir team. They were extremely helpful with the design & architecture. they have been valuable partners, and elixir continues to be an amazing language for tools like this and our Realtime server.
Upvote: | 149 |
Title: <a href="https://insightbase.ai" rel="nofollow">https://insightbase.ai</a> allows you to chat to your database and build analytics using the answers.<p>You can ask for anything and it will answer with a number a text or a graph.<p>You can build dashboards with widgets that auto-refresh at specified intervals.<p>This saves a lot of time and has the advantage that anyone can use it without knowing the database structure & coding.<p>What do you think about this?
Upvote: | 58 |
Title: I have not found any good description of the deal yet. Are the terms public?
Upvote: | 43 |
Title: Original article: https://www.svt.se/kultur/spotify-tystar-fore-detta-anstallda-i-uppsagningskontrakt<p>Google translation:<p>The expert: Spotify is trying to silence former employees in severance contracts
PUBLISHED APRIL 5, 2023
Spotify requires in the termination contract that former employees do not speak "disparagingly" about the company after termination of employment.<p>Now the streaming giant is accused of trying to intimidate former employees into silence.<p>- It is the only interpretation that is reasonable, says labor law expert Tommy Iseskog.<p>In January, Spotify laid off 6 percent of its employees. SVT has interviewed several of these, all of whom have signed a termination contract with a so-called loyalty clause.<p>It states that the dismissed person may not "speak disparagingly" about the company or former colleagues, that the provisions on confidentiality and fines stated in the employment agreement "apply after the end date" and that a written approval from Spotify is required to be able to talk about the dismissal with Spotify's customers and business partners.<p>According to a former employee, Spotify wants to scare people into silence.<p>- They may know that it is not legal, but it is still good to try to get people to voluntarily not talk about this externally.<p>Expert: Intimidation Tactics
According to Tommy Iseskog, labor law expert, the wording is unusual in Sweden - and legally unsustainable.<p>- Maybe you want to scare people and make people believe that it is legally binding, but that is just nonsense.<p>Can it be interpreted as Spotify wanting to silence its employees?<p>- Yes, that is probably the only interpretation that is reasonable, he says.<p>Several labor law experts confirm to SVT that the wording is unusual in Sweden and would not hold up under legal scrutiny.<p>Affects freedom of expression
Demanding silence after termination of employment is, according to Tommy Iseskog, an "Americanization" that mainly occurs in foreign companies operating in Sweden. He believes that in the long run it can have consequences:<p>- It is an attempt to restrict the social freedom of expression that we have here.<p>In the loyalty clause, Spotify states that the contract may be made public in special cases, for example to obtain the right to unemployment compensation, in courts, for unions or when required by law.<p>The streaming giant declines a taped interview, but responds in writing:<p>Spotify complied with all legal requirements and provided affected employees with a generous severance package and career support.
Upvote: | 75 |
Title: Working on a ML project and looking for an open source LLM that can be used in a commercial environment. As far as I'm aware, products cannot be built on LLAMA.<p>I don't want to use GPT since the project will be using personal information to train/fine tune the models.
Upvote: | 184 |
Title: What is your experience with layoffs? Did it affect your psyche, and did it change your commitment to potential future employers?<p>How you were able to overcome this situation emotionally and psychological?<p>Share your honest opinion
Upvote: | 40 |
Title: Run your scripts with Wolverine and when they crash, GPT-4 edits them and explains what went wrong. Even if you have many bugs it will repeatedly rerun until it's fixed.<p>Demo video on twitter: <a href="https://twitter.com/bio_bootloader/status/1636880208304431104" rel="nofollow">https://twitter.com/bio_bootloader/status/163688020830443110...</a>
Upvote: | 43 |
Title: Hi I am a blogger working on a small niche . I have written all these articles from the ground up with considerable effort. I don't want this to end up just as a free training data set for LLMs. Is there anything I can do to prevent that and still keep my site open free for visitors?
Upvote: | 46 |
Title: GitHub: <a href="https://github.com/EqualTo-Software/stackedit-sheet-markup">https://github.com/EqualTo-Software/stackedit-sheet-markup</a>
Upvote: | 120 |
Title: As the former CTO of an Insurtech and Fintech startup I always had the “pleasure” to keep regulators and auditors happy. Think of documenting who has access to what, quarterly access reviews, yearly audits and so on…<p>Like many others we couldn’t justify the Enterprise-plan for every SaaS tool to simply get access to SSO and SCIM/SAML APIs. For Notion alone the cost would have nearly doubled to $14 per user per month. That’s insane! Mostly unknown to people, SSO Tax also limits access to APIs that are used for managing user access (SCIM/SAML).<p>This has proven to be an incredibly annoying roadblock that prevented me from doing anything useful with our user data:
- You want to download the current list of users and their permissions? Forget about it!
- You want to centrally assign user roles and permissions? Good luck with that!
- You want to delete user accounts immediately? Yeah right, like that's ever gonna happen!<p>It literally cost me hours to update our access matrix at the end of every quarter for our access reviews and manually assigning user accounts and permissions.<p>I figured, there must be a better way than praying to the SaaS gods to miraculously make the SSO Tax disappear (and open up SCIM/SAML along the way). That’s why I sat down a few weeks ago and started building OpenOwl (<a href="https://github.com/AccessOwl/open_owl">https://github.com/AccessOwl/open_owl</a>). It allows me to just plug in my user credentials and automatically download user lists, including permissions from SaaS tools.<p>Granted, OpenOwl is still a work in progress, and it's not perfect. At the moment it's limited to non-SSO login flows and covers only 7 SaaS vendors. My favorite part is that you can configure integrations as “recipes”. The goal was for anybody to be able to add new integrations (IT managers and developers alike). Therefore you ideally don’t even have to write any new code, just tell OpenOwl how the new SaaS vendor works.<p>What do you think? Have you dealt with manually maintaining a list of users and their permissions? Could this approach get us closer to overcoming parts of the SSO Tax?
Upvote: | 220 |
Title: I’m Yousif, founder/CTO of OASIS AI, a new iOS app to elevate your communication skills.<p>https://apps.apple.com/us/app/oasis-ai/id1668222944<p>OASIS AI transforms natural speech into polished writing—college essays, business emails, social media posts, journals, and more. Everything is instant, automatic and efficient. and efficiency. It’s very simple.<p>Let’s face it: humans communicate constantly, but we suck at it.<p>We want to use LLMs to go beyond Grammarly's error correction and Google Translate's word-for-word translation. Instead, OASIS refines your thoughts and intentions, choosing the perfect words to express your ideas in any format or language. The app gives you a first draft, a starting place, and then you’re in control from there.<p>OASIS started off as English–to-English, and we are now in the process of rolling out language translation and speech transcription for over 30 languages, including Chinese, Japanese, Spanish, and Arabic. As a preview, the app enables English speakers to craft WeChat posts that mimic the casual writing of a native Mandarin Chinese speaker.<p>What do you think? What features should we add? How can we be better?<p>https://theoasis.com/
Upvote: | 103 |
Title: Former Apple/Facebook engineer, been leading engineering and product teams at startups the last few years. Have normally found a new job within a few weeks, but this time it has been 4 months.<p>Submitted 150 job applications last week. Got one interview with a recruiter. Had a few interview rounds over the last few months through old coworkers, they all lasted several months with long pauses - one still going 4 months in.<p>How is it going for everyone else searching right now? Is it just me?
Upvote: | 209 |
Title: The course topics can be anything.
Upvote: | 399 |
Title: A couple of days ago I was trying to check my messages on a real estate website (I won't say which). But scrolling only worked in landscape mode.<p>Then I tried to watch a TikTok that a friend sent me, but the website took forever to load. When it finished loading, the player didn't work.<p>What is happening? Is software quality in decline?
Upvote: | 47 |
Title: Hi HN,<p>As an indie hacker, the new Google Analytics (GA4) coming motivated me to look for a straightforward alternative that would also be affordable. I had a few basic product requirements and didn’t want to spend too much to replace a free product. There are a lot of great Google Analytics alternatives out there, but the pricing didn’t seem right. As someone who likes to just build things, many of which aren’t businesses yet, it didn’t make sense to pay for options like Plausible and Fathom out of the gate.<p>So I joined with a friend to build Beam Analytics. Beam gives you all the standard web analytics. It also comes with easy to create funnels so you can see how users move through your site. And we have a great proxy for cohort retention that doesn’t need you to log any data with us. It’s cookie-less and GDPR compliant.<p>The free tier is 100k page views per month so hopefully you’ll give it a try. There’s also a Wordpress integration to make integrating with WordPress sites as easy as a single click - <a href="https://wordpress.org/plugins/beam-analytics/" rel="nofollow">https://wordpress.org/plugins/beam-analytics/</a>.<p>Appreciate your feedback. You can also email us at hi (at) beamanalytics.io or DM me on twitter @TheBuilderJR.
Upvote: | 124 |
Title: "After Dark" by Berkeley Systems was a collection of screensavers for Macintosh and Windows from early 90s; "Flying Toasters" is probably the most famous and loved among them.<p>Using my very limited knowledge of C and X11 programming, I recreated it as a XScreensaver module for Linux.
Hope you find it useful or at least amusing!
Upvote: | 80 |
Title: Hi HN,<p>Today we’re launching GPT-4 answers on 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. Unlike vanilla GPT-4, Phind feeds in relevant websites and technical documentation, reducing the model’s hallucination and keeping it up-to-date. To use it, simply enable the “Expert” toggle before doing a search.<p>GPT-4 is making a night-and-day difference in terms of answer quality. For a question like “How can I RLHF a LLaMa model”, Phind in Expert mode delivers a step-by-step guide complete with citations (<a href="https://phind.com/search?cache=0fecf96b-0ac9-4b65-893d-8ea5708db222">https://phind.com/search?cache=0fecf96b-0ac9-4b65-893d-8ea57...</a>) while Phind in default mode meanders a bit and answers the question very generally (<a href="https://phind.com/search?cache=dd1fe16f-b101-4cc8-8089-ac56defb228d">https://phind.com/search?cache=dd1fe16f-b101-4cc8-8089-ac56d...</a>).<p>GPT-4 is significantly more concise and “systematic” in its answers than our default model. It generates step-by-step instructions over 90% of the time, while our default model does not.<p>We’re particularly focused on ML developers, as Phind can answer questions about many recent ML libraries, papers, and technologies that ChatGPT simply cannot. Even with ChatGPT’s alpha browsing mode, Phind answers technical questions faster and in more detail.<p>For example, Phind running on “Expert” GPT-4 mode can concisely and correctly tell you how to run an Alpaca model using llama.cpp: (<a href="https://phind.com/search?cache=0132c27e-c876-4f87-a0e1-cc48f07ccc20">https://phind.com/search?cache=0132c27e-c876-4f87-a0e1-cc48f...</a>). In contrast, ChatGPT-4 hallucinates and writes a make function for a fictional llama.cpp.<p>We still have a long way to go and would love to hear your feedback.
Upvote: | 1401 |
Title: I have been a Java Developer then full stack developer all my career. I want to get started on AI. I am not sure where to start as it is overwhelming to look at the rapid changes coming in. Are there a list of books i can start with? My goal is to be not outdated in 6 months and lose my job to chatGPT or some other version of it. So i asked ChatGPT to give me a detailed plan and here is what it gave me. Could you please let me know if this is a way to go or if there is a better way<p>Transitioning from a full-stack engineer to the AI field can be a fulfilling and exciting journey. To make the most of this transition, you should focus on building a strong foundation in mathematics, programming, and machine learning concepts. Here's a detailed learning plan, broken down into four stages:<p>Stage 1: Building a Strong Foundation<p>Mathematics:<p>Linear Algebra: "Linear Algebra and Its Applications" by Gilbert Strang<p>Calculus: "Calculus: Early Transcendentals" by James Stewart<p>Probability & Statistics: "Probability and Statistics for
Engineers and Scientists" by Ronald E. Walpole, Raymond H.
Myers, Sharon L. Myers, and Keying E. Ye<p>Optimization: "Convex Optimization" by Stephen Boyd and
Lieven Vandenberghe<p>Programming:<p>Python: "Python Crash Course" by Eric Matthes<p>Git: "Pro Git" by Scott Chacon and Ben Straub<p>Machine Learning & Data Science Basics:<p>"Introduction to Data Science" by Laura Igual and Santi Seguí
"Python Data Science Handbook" by Jake VanderPlas<p>Stage 2: Learning Machine Learning and Deep Learning<p>Online Courses:<p>Coursera: "Machine Learning" by Andrew Ng<p>Coursera: "Deep Learning Specialization" by Andrew Ng<p>Fast.ai: "Practical Deep Learning for Coders"<p>Books:<p>"Pattern Recognition and Machine Learning" by Christopher M. Bishop
"Deep Learning" by Ian Goodfellow, Yoshua Bengio, and Aaron Courville<p>Stage 3: Expanding Knowledge of AI Subfields<p>Natural Language Processing (NLP):<p>"Speech and Language Processing" by Daniel Jurafsky and James H. Martin<p>"Natural Language Processing with Python" by Steven Bird, Ewan Klein, and Edward Loper<p>Computer Vision:<p>"Computer Vision: Algorithms and Applications" by Richard Szeliski<p>"Deep Learning for Computer Vision" by Adrian Rosebrock
Reinforcement Learning:<p>"Reinforcement Learning: An Introduction" by Richard S. Sutton and Andrew G. Barto<p>"Deep Reinforcement Learning Hands-On" by Maxim Lapan<p>Stage 4: Staying Current and Gaining Practical Experience<p>Read research papers:<p>Subscribe to the arXiv mailing list in your areas of interest
Regularly read papers from conferences like NeurIPS, ICML, and ACL
Participate in online competitions:<p>Kaggle: Participate in machine learning competitions to improve your skills and build your portfolio
AIcrowd: Another platform for AI competitions
Contribute to open-source projects:<p>TensorFlow, PyTorch, or other popular AI frameworks
Look for projects on GitHub related to your interests and contribute by fixing bugs, implementing new features, or improving documentation
Network with AI professionals:<p>Attend AI conferences, workshops, and local meetups
Join AI-related forums, LinkedIn groups, or online communities like Reddit's r/MachineLearning
Remember that the learning plan can be adjusted based on your pace and interests. The key is to stay consistent and always be curious to learn more. Good luck on your journey into the AI field!
Upvote: | 74 |
Title: endpoints: /graphql /1.1 /2 /i
Upvote: | 183 |
Title: I think we’ve all heard quite a bit about how LLMs will impact SWE work in the near future, and would like to hear the other side. Please keep it realistic, no saying the job wont change dramatically or otherwise undersell the impact of LLMs.
Upvote: | 43 |
Title: Do you believe that GPT-4 will significantly reduce the demand for coders? If not, why not? I am curious to get your sense of this. I am not a professional coder. I write code for scientific work. But I am really curious.
Upvote: | 45 |
Title: As my daughter is on her school vacation would like to get some good reference of books, creative games I can introduce to her. I am more keen to get her know about art/programming/drawing/music/science<p>Any links where I can buy/download will be much appreciated, thanks in advance!
Upvote: | 149 |
Title: I plan on starting a side project but don't really have a niche yet.<p>I am interested in knowing what business you run is it a mobile app, website, Saas?<p>And how long it took you to reach $2k monthly revenue?
Upvote: | 530 |
Title: Hi HN community! I am excited to share my recently upgraded website, DebateDevil (<a href="https://debate-devil.com" rel="nofollow">https://debate-devil.com</a>), an AI chat platform designed to help users practice debates, challenge their ideas, and expand their thinking. I created this site to provide a more objective opinion and help users consider their hypotheses from multiple perspectives.<p>example discussin:
```
AI: The time has come to test your hypothesis and beliefs. I will argue against any topic. What do you want to debate?
You: AI will take away most jobs from humans.<p>AI: I understand your concern about AI taking away jobs from humans. However, it's important to note that AI is not designed to replace humans but to augment and enhance human capabilities. AI can handle repetitive and mundane tasks, freeing up humans to focus on more complex and creative tasks that require empathy, critical thinking, and problem-solving skills. Additionally, the increased efficiency and productivity brought by AI can lead to the creation of new job opportunities in areas such as AI development and maintenance.<p>You: While it's true that AI aims to augment human capabilities and increase efficiency, we cannot ignore the potential negative impacts on employment. Job displacement is a valid concern, especially for those in industries heavily reliant on routine tasks. Reskilling and upskilling may be necessary but can be challenging and costly. Additionally, new job opportunities in AI-related fields may not be accessible to everyone, leading to further social and economic disparities.<p>AI: I agree that job displacement is a valid concern. However, it's important to note that technology has always been changing the job market, and AI is just the latest technology to do so. While some jobs may become obsolete, new jobs will be created in other areas. For example, the rise of e-commerce has led to a decline in brick-and-mortar retail jobs but at the same time, created new job opportunities in online retail and logistics. As for the issue of reskilling and upskilling, there are various resources available such as online courses, training programs, and apprenticeships to help workers acquire new skills to adapt to changing job requirements. It's also important for governments and companies to invest in retraining programs to ensure that workers have access to these resources.<p>```<p>example judge
```
I think AI is superior in what you discussed; AI focuses on AI's ability to extend human capabilities and points out that new employment opportunities may be created due to the increased efficiency and productivity that AI brings. On the other hand, you are concerned about job losses due to AI, but AI points out that past technological innovations have also changed the employment landscape and that new job opportunities can be created. You also indicate that resources exist for retraining and upskilling to address job losses due to AI, and state that it is important for governments and businesses to invest in these programs.<p>```<p>I'm eager to hear what the HN community thinks! Please give it a try and let me know your thoughts, feedback, and suggestions for improvement. Thanks for checking it out!
Upvote: | 84 |
Title: While a thread of 2k+ MRR is going, I thought it'd be cool to hear from folks who aren't quite at 2k yet.
Upvote: | 95 |
Title: Connect two articles on Wikipedia, but do it the long way. I've always been a fan of the theory of six degree of separation, but it's an overused concept when exploring the Wiki-graph.<p>Instead of showing the shortest path, which in my opinion is "boring" and ends up connecting super-important central articles, I came up with my own method: WikiBinge selects the smaller, less represented articles on Wikipedia. In a WikiBinge path, the underdogs are the kings!<p>How does it work? It's pretty straightforward! Compute PageRank on the Wiki-graph and assign as weight of each edge the PageRank value of the destination node. A WikiBinge path is then simply a shortest path using these weights: the algorithm will then favor paths passing through articles with lower PageRank values.<p>More on the motives to build this here: <a href="https://www.jamez.it/project/wikibinge/" rel="nofollow">https://www.jamez.it/project/wikibinge/</a><p>This is an older project of mine, but it never got much exposure, so I'm humbly submitting it now.
Upvote: | 246 |
Title: A person's heart rate fluctuates based on factors such as activity, environment, and mental state. Additionally, heart rate variations occur on a much shorter timescale, with changes happening every few seconds. During inhalation, heart rate slightly increases, and it decreases again during exhalation, a phenomenon known as respiratory sinus arrhythmia. There are several reasons for this. Firstly, an increase in volume within the thoracic cavity (the space between the ribs) leads to a corresponding drop in pressure, which is counteracted by a rise in heart rate to regulate blood pressure. Secondly, breathing is deeply intertwined with the autonomic nervous system, which involves the opposing activation of the parasympathetic ("rest and digest") and sympathetic ("fight or flight") responses. By measuring heart rate variability and regulating our breath, we can gain insight into our stress response and learn to control it.<p>The Polar H10 heart rate monitor provides basic heart rate measurements from a chest strap ECG, compatible with a number of fitness tracking apps that monitor heart rate zones wirelessly using bluetooth. The device also provides raw sensor data including an accurate measure of the timing between heartbeats (interbeat interval, with millisecond resolution) and 200 Hz accelerometer data. In this project this information is used to measure breathing rate and heart rate variability, and explore the relationship between the two.
Upvote: | 51 |
Title: I'm interested in learning "modern" SEO and other growth hacking related topics as it is relevant in 2023 but want to steer clear of bullshit courses. What are your top 3 favourite resources?
Upvote: | 45 |
Title: Early 80s computers start up with a BASIC prompt and hence encourage you to learn programming right away.<p>Why don't smartphones do something similar?
Upvote: | 162 |
Title: Not being a hater since I also leverage open ai api heavily in my projects but to call it AI project would be silly.<p>It’s just like building on top of AWS or GCP.<p>Open AI is an AI company. Extending LLMs projects like memory for chat gpt etc are AI projects.<p>Some two minute UI that makes a call to open ai to show you “As a large language model …” is not impressive and NOT an AI company.<p>Isn’t it a lie basically to claim so? Or am I missing something.<p>In my case I’m building a dev tool that uses open ai product in the workflow. My product is NOT an AI company. It’s a dev tool for app developers.<p>ESL btw.
Upvote: | 78 |
Title: For the past hour I've been trying to cancel a Docusign subscription that I never even wanted in the first place.<p>In order to cancel my subscription I was forced to open a support case. After opening one, they're asking me for all sorts of weird things such as mandatory "feedback" and made-up "security" questions. I'm still struggling to cancel my subscription and starting to give up.<p>A truly unethical company with horrible practices.
Upvote: | 75 |
Title: I'd be curious to learn how effective has GPT4 been to enable product features and what it means for the things we might see in the future.<p>In particular, I have the following questions:<p>1. What was the product you were working on?<p>2. Were there any new software engineering challenges that came from working with GPT4 (e.g. output quality, testing, monitoring, etc.)?
Upvote: | 80 |
Title: I hope this doesn't entice a flamewar, but speaking honestly here, whenever I discuss this (PostgreSQL vs MySQL) with folks working in the data space, PostgreSQL is always touted as the superior RDBMS.<p>Now I know that MySQL is the most popular (because there's a banner on mysql.com that says so! /s), and I get the sense that on raw speed alone, it's probably faster (this is purely anecdotal), but folks that use PostgreSQL speak about it with an air of superiority.<p>So I'm hoping some of you HN users can shed light on this, based on your experience.<p>Is there a clear winner, or is it purely a case of "it depends"?<p>(and if it's the latter, is there a generally agreed comparison list of pros and cons that I can refer to?)<p>Thanks!
Upvote: | 83 |
Title: PrintableSearch is a new 3d printable search engine to find the top 3D printables from the best model sites. It differs from others by ranking models between sites accurately and it has a modern ad free interface.
Upvote: | 131 |
Title: I built this kubernetes operator as a proof of concept this weekend.. It only has a single required item in the spec, a freeform description field. The operator will use chatgpt to generate a spec, then immediately apply it to the cluster. It makes some attempt to correct errors if there's a problem with the syntax. It will leave additional comments, questions or instructions in the status field of the object. I built this in a weekend and it's still quite unrefined. It's in no way production ready, please don't use it for anything real, but it works better than you would think, considering how simple it is. If you're going to use it, run it on a local cluster like 'kind'.<p>Some descriptions to try:<p>* install a redis namespace with a redis cluster and a service in it
* create an argocd application in the argocd namespace to install velero.
* write a python script that lists all ec2 instances in us-east-1, and run it as a k8s job with the aws credentials already saved in the default namespace..<p>a somewhat longer description that also worked:
given the following spec:
---
kind: MagicHappens
apiVersion: gptmagic.io/v1
metadata:
name: foo
spec:
description: this is a freeform description field that will be sent to chatgpt to generate kubernetes resources
dryRun: false
---
Can you create more magic happens resources, each of which describes an argocd application that needs to be created to install a helm chart for one of the standard cluster addons that need to be installed on a cluster for it to be production ready. The description should be be freeform text like the following: "Create an argocd application in the argocd namespace to install istio from the helm chart with all the defaults" or "Create an argocd application in the argocd namespace to install prometheus and grafana, with an ingress enabled for grafana". Be very thorough and included as many apps that might be needed for a prod ready cluster using industry standard CNCF projects if possible.<p>(this produces a list of additional resources for the operator, which the operator then goes on to create argocd applications for -- it also left comments with instructions on one of the resources for how configure it to work with your cloud provider<p>something to note is that since you can run arbitrary containers with arbitrary commands, and chatgpt can write arbitrary code, you don't have to limit yourself to k8s stuff.. if you've got saas credentials on the cluster, you can just tell it to run a python script as a job to do whatever you want.<p>Since most people are cowards, there's a dryRun field that defaults to true, so it only attaches the spec to the object.<p>It is <i>scary</i> how well this works.
Upvote: | 48 |
Title: TL;DR: as a dev, what's the most valuable skill to learn when wanting to become a more well-rounded solo founder and how do you learn that skill the quickest?<p>Some thoughts (I could be entirely off-base, hence the TL;DR):
I have a side project in mind. I know nothing about marketing other than some basic ideas (I think I read half of the book Traction at one point?).<p>I was wondering, should I:<p>1. Learn marketing<p>2. Start a side project?<p>Goal: in 5 years from now, I want to be well positioned to launch a startup? What startup? I'm an idea guy by heart. I studied a master CS 5 years ago + did a software engineer career, in order to implement my ideas. In other words: I'll think of something that I find fun and will solve a problem.<p>Also, if learning marketing first is the smarter option. Should I learn marketing by taking a course, or should I take a side step (or step back) in my career and do a marketing role?
Upvote: | 55 |
Title: I graduated from college about decade+ ago, took few OS & systems programming courses, but didn't find them interesting back then. However, now, more than a decade later, I'm finding myself bored at my cookie cutter C#/Asp.net/SQL(+ cloud) job.
What would be your suggestions getting into OS/System programming in 2023 ?
For starters, I'm thinking about learning assembly, which I never learned.
Upvote: | 40 |
Title: In a world where nothing seems sure anymore what would you do? Would you bother getting an education? When knowledge becomes meaningless is there a reason to learn?
Upvote: | 61 |
Title: My main motivation was that I wanted to be able to drill down and filter across all the available stocks, look at the data for myself, and narrow down on the stocks I am interested based on my own sets of criteria, and make data-driven analysis for my personal investment strategies.<p>I used PostgreSQL as the backend database for ELT data pipelines, and used Citus Data cstore_fdw for columnar compression for the final dataset. All financial data is coming from SEC Edgar, <a href="https://www.sec.gov/developer" rel="nofollow">https://www.sec.gov/developer</a>. I used Python for downloading most of the data.<p>I also run the data load development locally on my home Ubuntu server that I built 5 years ago. I bought 4TB of M2 disks for best database performance, with PRIME B360M-A motherboard and Intel Chip Coffee Lake S.<p>I built the website simply using WordPress, and I run Apache Superset using gunicorn via Apache Webserver reverse proxy.<p>The registration form I had to build myself with PHP and some JavaScript, because it needed to automatically create a SuperSet user upon registration. Otherwise, I would need to input everyone manually. I used Python again for the data integration.<p>Please don't use the database directly as an investment tool, as its in Beta, and the data still needs to undergo heavy data quality checks, please confirm all the numbers yourself, as I provide a link for every company to the SEC filings.
Upvote: | 113 |
Title: Hey, Jared Palmer (creator of this playground) here. Really excited to ship this.
I’ve been building this over the past few weeks to compare LLMs from different providers like OpenAI, Anthropic, Cohere, etc. At Vercel, I manage our Frameworks division (including Next.js, Svelte, and Turbo) and wanted to also dogfood some of the latest features in a slightly larger application.
This playground takes a lot of inspiration from <a href="https://nat.dev" rel="nofollow">https://nat.dev</a> and is built on Tailwind, ui.shadcn.com, and some upcoming Vercel products we’re announcing soon. We’re going to continue adding models to compare and add other frameworks to generate code snippets from.
Upvote: | 249 |
Title: Currently performing searches. Sometimes doesn't work at all. Searches now seem to display cards when it works properly instead of urls with text.
Upvote: | 179 |
Title: I tried to search some random stuffs but it's not showing results. In Canada.<p><i>Edit</i>: Back around 9:30pm EST
Upvote: | 102 |
Title: Hey HN!<p>I’m Gal, co-founder at Checksum (<a href="https://checksum.ai" rel="nofollow">https://checksum.ai</a>). Checksum is a tool for automatically generating and maintaining end-to-end tests using AI.<p>I cut my teeth in applied ML in 2016 at a maritime tech company called TSG, based in Israel. When I was there, I worked on a cool product that used machine learning to detect suspicious vehicles. Radar data is pretty tough for humans to parse, but a great fit for AI – and it worked very well for detecting smugglers, terrorist activity, and that sort of thing.<p>In 2021, after a few years working in big tech (Lyft, Google), I joined a YC company, seer W21, as CTO. This is where I experienced the unique pain of trying to keep end-to-end tests in a good state. The app was quite featureful, and it was a struggle to get and maintain good test coverage.<p>Like the suspicious maritime vehicle problem I had previously encountered, building and maintaining E2E tests had all the markings of a problem where machines could outperform humans. Also, in the early user interviews, it became clear that this problem wasn’t one that just went away as organizations grew past the startup phase, but one that got even more tangled up and unpleasant.<p>We’ve been building the product for a little over a year now, and it’s been interesting to learn that some problems were surprisingly easy, and others unusually tough. To get the data we need to train our models, we use the same underlying technology that tools like Fullstory and Hotjar use, and it works quite well. Also, we’re able to get good tests from relatively few user sessions (in most cases, fewer than 200 sessions).<p>Right now, the models are really good at improving test coverage for featureful web-apps that don’t have much coverage (ie; generating and maintaining a bunch of new tests), but making existing tests better has been a tougher nut to crack. We don’t have as much of a place in organizations where test coverage is great and test quality is medium-to-poor, but we’re keen to develop in that direction.<p>We’re still early, and spend basically all of our time working with a small handful of design partners (mostly medium-sized startups struggling with test coverage), but it felt like time to share with the HN community.<p>Thanks so much, happy to answer any questions, and excited to hear your thoughts!
Upvote: | 78 |
Title: Ideally I want to contain and run LLM output of my domain-specific language, but it seems that I would need to fine-tune existing models. What’s the easiest online or local solution?<p>How to automatically generate:
a broad array of security tests;
the most efficient code;
the most readable and extensible code
Upvote: | 70 |
Title: Just paste in a chunk of systemd (or whatever) logs and start asking questions. Often just pasting in the logs and pressing enter results in it identifying potential problems and suggesting solutions. It helped me troubleshoot a huge amount of issues on linux desktops and servers that would have taken me a lot longer with google - even if it doesn't always give the right solution, 99% of the time it at least points to the source of the error and gives me searchable keywords.<p>Note it works much better with GPT4 - gpt3.5 tends to hallucinate a bit too often.
Upvote: | 256 |
Title: I ran Legit Torrents for ~17 years and shut it down recently. The homepage is now a nostalgic look back at that time.
Upvote: | 749 |
Title: How did you feel then vs now? Are you doing alright?<p>Are you looking for/have you found work? How do pay/benefits (remote, etc.) compare?<p>How have your feelings about the industry changed?
Upvote: | 104 |
Title: Hi HN! We're thrilled to share CozoDB v0.6, a monumental update to our FOSS database, which already unifies relational and graph features. With the addition of vector search, CozoDB becomes an even better companion for LLMs like ChatGPT.<p>This release introduces vector search using HNSW indices within Datalog, enabling seamless integration with powerful features such as ad-hoc joins, recursive Datalog, and classical whole-graph algorithms. This update significantly broadens CozoDB's capabilities.<p>Check out the linked release note for an in-depth look at the new features, comparisons to other systems, and intriguing AI development possibilities. We'd love for you to take a look! I'll be here to answer any questions you might have.<p>Looking forward to your feedback!
Upvote: | 137 |
Title: Hi everyone,<p>I just launched Bank List, an online banking directory to help people find the perfect bank for their needs.<p>It currently has 170+ banks/EMIs and it's mostly focused on individuals and businesses from Europe, but I plan on adding new regions soon.<p>The business model is affiliate marketing, I earn commission when someone signs up on any of these banks through my link. Maybe I will also add featured listings on the site, still need to think about this though.<p>I would love to receive feedback about the website and business idea in general.<p>Thank you
Upvote: | 50 |
Title: Hey hn! I'm one of the co-founders of highlight.io, an open source monitoring tool.<p>Today we're sharing a ClickHouse-based logging solution we've been working on. We wanted to showcase how we built it and share how you could try it out to give feedback.
Since we started working on highlight.io, we've been hyper-focused on "cohesion", or ensuring that when you install your monitoring stack, all of the resources in that stack (user interactions, requests, traces, logs, etc.) are connected in a consumable way. We've written up more about our philosophy on this here [1].<p>We started building towards this by connecting your client-side app and your server-side exceptions with session replay and exception monitoring; i.e. if an error happened in a server-side app, we would make it easy (with session replay) to trace all the steps that a user took leading up to it.<p>Especially for larger companies using highlight.io, the request to tie in logs came up repeatedly, and we wanted to build this with the same philosophy in mind. Now, you'll see client-side and server-side logs all in one place, brought together in the context of a user session, as well as logs in the context of an error.<p>Like the rest of our stack, this project is written in Go and Typescript, and for log ingestion/querying, we're using ClickHouse [2]. Before deciding on ClickHouse, we were planning to use OpenSearch (an aws fork of elasticsearch [3]) for this part of our product, but as our traffic has increased, we encountered quite a few pains with write throughput for OpenSearch. After evaluating a few options, we eventually landed with ClickHouse (their cloud offering was icing on the cake), which has also proven to be much more cost-effective so far.<p>Building with ClickHouse from scratch has been an exciting journey. Eric (the mastermind behind this project) wrote a blog post [4] on a handful of ClickHouse learnings we've gathered since starting the project.<p>For those wanting to try out the product locally, you can run the following commands [5]:<p>git clone --recurse-submodules <a href="https://github.com/highlight/highlight">https://github.com/highlight/highlight</a>
cd highlight/docker;
./run-hobby.sh;<p>To send logs to highlight, you can use your own OpenTelemetry implementation [6] or use our SDKs [7] which provide lightweight wrappers over OTEL.<p>Like the rest of highlight.io, we plan to make money from this with our hosted cloud offering. For those interested in trying out the cloud-hosted version, you can get setup at app.highlight.io.<p>To open the floor for feedback, we would love to get some thoughts on what we've built so far. Beyond that, what are parts of a logging product you wish you had with your current setup? And are there any notable pain-points of using a hosted monitoring product? (We're toying with the idea of an enterprise deployment). Excited to hear from everyone.<p>[1]: <a href="https://highlight.io/docs/general/company/product-philosphy">https://highlight.io/docs/general/company/product-philosphy</a><p>[2]: <a href="https://clickhouse.com" rel="nofollow">https://clickhouse.com</a><p>[3]: <a href="https://news.ycombinator.com/item?id=26780848" rel="nofollow">https://news.ycombinator.com/item?id=26780848</a><p>[4]: <a href="https://www.highlight.io/blog/how-we-built-logging-with-clickhouse">https://www.highlight.io/blog/how-we-built-logging-with-clic...</a><p>[5]: <a href="https://www.highlight.io/docs/getting-started/self-host/self-hosted-hobby-guide">https://www.highlight.io/docs/getting-started/self-host/self...</a><p>[6]: <a href="https://www.highlight.io/docs/getting-started/backend-logging/http">https://www.highlight.io/docs/getting-started/backend-loggin...</a><p>[7]: <a href="https://www.highlight.io/docs/getting-started/overview#for-your-backend-logging">https://www.highlight.io/docs/getting-started/overview#for-y...</a>
Upvote: | 192 |
Title: Hey HN, I hacked together this fun project that lets you play any article on the front page of Hacker News as an interactive text game.<p>You can create a game about anything by going to infinityarcade.com/<anything you want goes here><p>You can also use a URL, which is what this page uses.<p>The whole thing is open source: <a href="https://github.com/themaximal1st/InfinityArcade">https://github.com/themaximal1st/InfinityArcade</a>
Upvote: | 109 |
Title: Tweet by creator:
<a href="https://twitter.com/willdepue/status/1649147091573432321" rel="nofollow">https://twitter.com/willdepue/status/1649147091573432321</a>
Upvote: | 154 |
Title: Spent the past few days making this - something halfway between a demoscene program and an experimental film. I wanted to celebrate the unique computer-y aesthetics of flash, while showing off some weird and obscure tricks I've picked up over the years since it's been deprecated. (Also, some maybe-not-subtle commentary about AI-art and the tools of the future)
Upvote: | 267 |
Title: Using threaded emscripten to speed up the generation and offload the main loop. No SIMD or other optimizations. Might work faster with #enable-experimental-webassembly-features enabled.<p>Tested in x86 Chrome and Firefox, Apple Silicon Safari<p>Run it yourself: <a href="https://github.com/lxe/ggml/tree/wasm-demo">https://github.com/lxe/ggml/tree/wasm-demo</a><p>Thanks, <a href="https://github.com/ggerganov/ggml">https://github.com/ggerganov/ggml</a>,
Upvote: | 185 |
Title: To the Muslims on HN, Eid Mubarak! And to everyone else, Eid Mubarak!<p>For those who don't know. Eid is a day of celebration after the month of Ramadan, in which Muslims fasted for 30 days from sunrise to sunset with no food or water. It's something 2B people around the world celebrate to today or tomorrow (moon sighting permitted).<p>A note on Ramadan. To those interested in intermittent fasting, longevity, and coming back to a more human experience not drowning in technology, food and consumerism I would say check it out! After over 20 years of doing it I'm still learning something new every year, or I should say, unlearning bad habits we've created for ourselves as a society through abundance.<p>Hope you all have a great day!
Upvote: | 815 |
Title: Recently, I write an interpreter for subset of Python using Go. I haven't seen source code of CPython, all is just written according to my understanding of Python so there may be some unusual things. Hope you can play fun with it. Any suggestion is welcome, glad to get some feedback.
Upvote: | 65 |
Title: Hi HN! We’re Ted and Thomas from Bend (YC S22). Today we’re launching a modern corporate card built for climate-friendly business.<p>Just by shifting some of your company spending to Bend, you’ll be helping to reduce climate-changing emissions, thanks to the carbon removal rewards you’ll automatically earn, and our embedded climate analytics. But we know that signing up for a new financial service can take some upfront effort, so we wanted to sweeten the deal — create an account today (Friday Apr 21), and we’ll deposit a *$100 gift* in your account to start spending immediately — see the site for details.<p>How it works:<p>- Bend is a modern card and cash account, combined with free carbon accounting and climate rewards<p>- Bend works similarly to how you’d use a business bank account. Send and receive ACH and wires, earn 3% yield on deposits, issue physical and virtual cards with budgets, etc.<p>- Carbon accounting typically runs companies $40k/year and up. We make money off the card interchange (same as any bank), enabling us to offer the carbon accounting for free, and pass through carbon removal rewards.<p>- The way we actually do the carbon accounting is laid out here — we always return a confidence score with every CO2e assessment: <a href="https://usebend.com/how-it-works">https://usebend.com/how-it-works</a><p>- We have awesome customers like Harmonic AI <a href="https://bend.green/harmonic" rel="nofollow">https://bend.green/harmonic</a> and Fathom <a href="https://bend.green/fathom" rel="nofollow">https://bend.green/fathom</a>. If you choose, you can create a report of your climate action through Bend like these ones, which you can share with customers, employees, investors, etc.<p>Why tracking and reducing climate-changing emissions is worthwhile:<p>- It’s good for the planet (and free!)<p>- It will help you sell your products to B2C customers who care about the climate, and B2B customers who have set net zero goals (64% of global market cap companies now report on climate data, and are increasingly asking their suppliers to do the same)<p>Our stack:<p>- Rust back end<p>- NextJS front end<p>- Hosted on Vercel<p>- Database from PlanetScale<p>- Card issuing and treasury powered by Stripe and partner banks<p>- Charts with Visx<p>How we got here:<p>We launched Bend last summer as an API <a href="https://news.ycombinator.com/item?id=32166140" rel="nofollow">https://news.ycombinator.com/item?id=32166140</a>. We’re now launching cards directly, because we’re able to offer the functionality for free, and it’s easier for startups to try it out. Prior to Bend, Thomas and I worked at Abacus, where I was one of the co-founders <a href="https://news.ycombinator.com/item?id=7219025" rel="nofollow">https://news.ycombinator.com/item?id=7219025</a> — our background is in fintech, and our goal is to layer in climate insights on top of your financial stack.<p>We’d love it if you tried Bend out!
Upvote: | 43 |
Title: Hi All, since I am consistently overwhelmed with all the all the ai/gpt news on the hn front page I built this simple extension to filter out such articles and show them on the extension popup if I'm in the mood to read. I am interested in the topic but just tired of reading the similar discussions (local models, llama on mac, ai taking away our jobs, etc) all the time. Every few days I search for topics on hn.algolia.com and go over any exciting development.<p>Ironically, since this was the first time I built an extension chat-gpt helped build 90% of the extension since . Also once installed, this show hn will disappear from your feed too :)<p>Firefox : <a href="https://addons.mozilla.org/en-US/firefox/addon/hacker-news-filter/" rel="nofollow">https://addons.mozilla.org/en-US/firefox/addon/hacker-news-f...</a><p>Github : <a href="https://github.com/prashantkhurana/Hacker-News-Filter-AI">https://github.com/prashantkhurana/Hacker-News-Filter-AI</a>
Upvote: | 99 |
Title: I am working on a modular open source framework called Griptape that allows Python developers to create LLM pipelines and DAGs for complex workflows that use rules and memory.<p>Griptape can be thought of as "Airflow for LLMs," providing an alternative to the agent-based LangChain approach.<p>Developers can also build reusable LLM tools with explicit JSON schemas that can be executed in any environment (local, containerized, cloud, etc.) and integrated into Griptape workflows. They can also be easily converted into ChatGPT Plugin APIs and LangChain tools via adapters.<p>Tools can be thought of as any executable code that allows LLMs to interact with the outside world (via ReAct and Toolformer techniques): email, docs, spreadsheets, Jira tickets, web pages/search, etc.<p>The best part about tools is that they can be executed in isolated environments, significantly reducing potential security risks associated with running LLM-generated code and API calls.<p>What do you think? What are some of the use cases that you have in mind for reusable tools?
Upvote: | 48 |
Title: Hello all,<p>This is my second year of working on a project[1] with the goal of browsing the web, on an Android smartphone, without reliance on Wi-Fi or mobile data. While this concept might seem aimless, my goal was to provide a way for people in areas with limited, expensive, or censored cellular internet access a way to view the web in a basic format. I finished work on a basic client-server model last year[2], and this year, I implemented a new pseudo-distributed peer-to-peer model that allows any TxtNet Browser user to use their own smartphone to run a background server service that communicates via the user's own primary mobile number. The main advantage to this model over last year's use of the Twilio API is the fact that with an unlimited SMS plan from a consumer carrier, you will likely end up paying significantly less than the amount you would pay for Twilio credits (averaging about ~$0.50 per website). There's a lot going on with the stateless nature of SMS, GSM-7 encoding, and Brotli compression, so please ask any questions you might have!<p>I've also started up a test server instance running on a +1 country code phone number, so feel free to test out the app with your own smartphone. Like mentioned in the GitHub repo, please be aware that I (necessarily) have access to every phone number and associated request that is sent. Of course, anyone can host their own server instance, and if you would like to share it, feel free to get in touch so I can add the number to the repo! Also, there are likely many bugs still lurking, so feel free to report those.<p>[1] <a href="https://github.com/lukeaschenbrenner/TxtNet-Browser/">https://github.com/lukeaschenbrenner/TxtNet-Browser/</a><p>[2] <a href="https://news.ycombinator.com/item?id=32905496" rel="nofollow">https://news.ycombinator.com/item?id=32905496</a>
Upvote: | 234 |
Title: They depend on ASML lithography machines, right? Intel, Samsung and other foundries could/do also buy ASML's machines, so what gives TSMC their advantage?
Upvote: | 149 |
Title: Is prompt engineering a snake oil? The engineering, as the word stands so far in history is about observing clear limitations and capabilities of something and then "engineering" things around it such as "engineering a compiler" is all about knowing the underlying processor, memory and other characteristics and then engineering a solution that converts a text notation to a stream of instructions that try to strike a balanced tradeoff.<p>But with LLMs, no one knows their workings once they are trained. The other day, Bark model[0] for text to speech, the team itself has following to say on details:<p>> Below is a list of some known non-speech sounds, but we are finding more every day. Please let us know if you find patterns that work particularly well on Discord! [laughter], [laughs], [sighs]....<p>So that's the team themselves not knowing what their model is capable of then how come prompt engineering is any engineering at all?<p>[0] <a href="https://github.com/suno-ai/bark">https://github.com/suno-ai/bark</a>
Upvote: | 199 |
Title: I wonder why I barely see this option on websites altough it seems so perfect to me. Is the WebAuthn protocol complicated to implement or are there any other downsides I might overlook?
Upvote: | 112 |
Title: I really don't like any changes to my OS (and software) once set up, unless I choose to install some change. I understand the need for security updates but I wish they were not bundled with features, i.e. how Windows 10 updates everything compared to Windows 7 specific update selection.<p>So, I migrated from Windows 7 to Ubuntu, but now Ubuntu is down the same path where everything changes and breaks by itself all the time.<p>Where should I migrate next? How is Mac with updates; is security tradeoff better when staying on old major versions? What flavor of Linux is most like Ubuntu before snaps?
Upvote: | 48 |
Title: For the past several months I have been getting hit with several card testing attacks. I sell a product with a subscription (which is processed via Stripe). The problem is that everytime this occurs I have to manually go into my account and determine if it is a card testing attack and then if it is then I have to refund and block the payment. Not a big deal if it happens once or twice but has been happening very frequently on my account. I've tried everything I can on my end but it's very limited what I can do since they don't even go through my website (I use Stripe Checkout, which they say is excellent protection against card attacks - which seems untrue to me). They simply get my public key (which I've rotated) and create their own checkout session. I've reached out to Stripe SEVERAL times and it is really annoying that they are unable to stop it.<p>What can I do, short of switching to PayPal or another provider?<p>Thanks!
Upvote: | 83 |
Title: RATH is an open-source software for data exploration and visualization. It automates your workflow in data analysis. You can give RATH a CSV file and it will generates visualizations with insights and patterns automatically. It also has a data painter tool for more casual interaction with charts (<a href="https://youtu.be/djqePNyhz7w" rel="nofollow">https://youtu.be/djqePNyhz7w</a>). It also has a lot of cool features like autovis, causal discovery, GPT integration, predictive interaction for text feature extraction, etc. Some of RATH’s components are also independent open-source project you might be interested:<p><pre><code> Like
</code></pre>
+ Graphic Walker (<a href="https://github.com/Kanaries/graphic-walker">https://github.com/Kanaries/graphic-walker</a>): A lite embeddable component for visual analysis.<p>+ PyGWalker (<a href="https://github.com/Kanaries/pygwalker">https://github.com/Kanaries/pygwalker</a>): turning your pandas dataframe into a Tableau-style User Interface for visual exploration.<p>RATH is a collection of interesting ideas that we think the next generation of data analysis software should be, so there might be many features that not well organized to be a united app. Tell me which feature you prefer and which is not. Looking forward for your ideas and advice.
Upvote: | 61 |
Title: (This was previously submitted as <a href="https://news.ycombinator.com/item?id=2246856" rel="nofollow">https://news.ycombinator.com/item?id=2246856</a>)<p>The Ravenna Ultra-Low-Altitude Vehicle is a backyard rocketship treehouse nestled in the Seattle neighborhood of Ravenna. Click the link to see a demo video (<a href="http://rocket.jonh.net" rel="nofollow">http://rocket.jonh.net</a>).<p>The hexagonal treehouse is about 6.5 feet (2 meters) across at its widest point. The frame is welded mild steel with riveted aluminum siding. It contains nearly 800 LEDs forming dozens of numeric displays spread across 14 control panels, each with an acrylic face laser-cut and etched with labels such as "Lunar Distance" and "Hydraulic Pressure". The pilot controls the rocket using a joystick and panels full of working switches, knobs and buttons. Underneath the capsule are three "thrusters" that shoot plumes of water and compressed air under the control of the pilot's joystick, simulating real positioning thrusters. Takeoff and docking sequences are augmented by a paint-shaker that simulates the vibration of a rocket engine. Sound effects complete the illusion, with a powered subwoofer that gives the rocket a satisfying rumble.<p>When it was built in 2011, rocket operations were controlled by three Atmega328 microprocessors on custom-fabricated printed circuit boards, running a small operating system, RULOS, built just for this project. A trench running from the house to the rocket carries 12VDC power for the lighting and electronics, water for the thrusters, compressed air, and several data signals.<p>Since 2011, the two-person team has upgraded it, here is a recent update from the makers:<p>One of the most visible changes is replacing the primary 4-line display with a slicker 6-line display (i.e., 6 rows of 8 columns of 7-segment LEDs). The audio synthesizer has been upgraded to a PCB that can generate 50khz, 16-bit audio. The interconnection bus, which had been flat IDC cable carrying individual on/off lines, was upgraded to a true I2C-based networked distributed system with over a dozen individually addressable targets, all interconnected by standard cat5 cable that carries both our I2C protocol and power. We also moved much of the electronics from 8-bit atmega328s to newer, 32-bit STM32F3's. RULOS has been expanded into a pretty general purpose embedded systems platform ported to 5 major lines of CPU (atmega, attiny, stm32, nxp lpc, and esp32). We've used it for dozens of other projects in the last 12 years, including a nanosecond-accurate timestamper, a GPS datalogger, an air quality sensor, various little electronic control boards for toys (e.g. these, and this), and an autonomous boat (that sank). It is all available on Github: <a href="https://github.com/jelson/rulos">https://github.com/jelson/rulos</a>.
Upvote: | 410 |
Title: I've had this IP for almost a decade. It's not on spamhaus, DKIM and SPF are correct, and I've signed up for MS's Junk Mail Reporting Program and Smart Network Delivery Services. They still reject my mail with<p>> Unfortunately, messages from [45.55.34.226] weren't sent. Please contact your Internet service provider since part of their network is on our block list (S3140). You can also refer your provider to http://mail.live.com/mail/troubleshooting.aspx#errors.<p>I went through their support channels and they were completely useless.<p>> Thanks for your patience while we investigated your request.<p>> Below your IP address(es) and their status(es) are listed.<p>> Not qualified for mitigation<p>> 45.55.34.226;<p>> The IP(s) above do not qualify for mitigation.<p>> Please note: This outcome indicates behavior that misses standards; please review Improving E-mail Deliverability into Windows Live white paper for helpful tips.<p>...<p>> What standards are missing? DKIM and SPF are passing and I got the IP taken off of Spamhaus recently. Other email servers like gmail aren’t finding an issue.<p>...<p>> Your IP (45.55.34.226) was blocked by Outlook.com because Hotmail customers have reported email from this IP as unwanted. One possible explanation for this is the automatic forwarding of unfiltered inbound messages, including unwanted messages, to Outlook.com/MSN addresses.<p>> Please confirm that your emails comply with Hotmail’s technical standards.<p>> For more detailed information about best sending practices to Outlook.com users, please review Outlook.com Enhanced Deliverability white paper.<p>...<p>> I’ve ensured there is no open relay, and I’ve only tried to send messages to my own Hotmail address so far. There are no other user accounts on this email server, just me.<p>> I’ve also signed up for the Junk Mail Reporting Program/Smart Network Data Services with that domain/IP. I don’t see any incidents there.<p>> Around what day/time was email reported as wanted?<p>...<p>> Thank you for contacting the Outlook.com Deliverability Support Team.<p>> As previously stated, your IP (45.55.34.226) do not qualify for mitigation at this time. I do apologize, but I am unable to provide any details about this situation since we do not have the liberty to discuss the nature of the block.<p>> At this point, I would suggest that you review and comply with Outlook.com’s technical standards.<p>> We regret that we are unable to provide any additional information or assistance at this time.<p>I've gone through all the links they sent me and nothing is wrong with my email server. It's impossible that I could have ever sent spam. They just decided they don't like me for no reason and I don't get to send them mail.
Upvote: | 50 |
Title: Why?<p>- Decouple application deployment from releases<p>How it works:<p>- Make changes to your features and segments (YAMLs) via Pull Requests
- Generate datafile (JSON file) in CI/CD workflow and upload to your CDN
- Fetch datafile in your application runtime and consume with SDKs<p>Supports:<p>- Feature flags: boolean flags
- Experimentation: a/b tests
- Segments: targeting traffic with conditions
- Variables: namespaced under each feature and conditional
- Gradual rollouts: avoid big bang releases, go from 0% to 100%
- Consistent bucketing: same user sees same variation
- Multiple environments: prod, staging, test, and more
- Force bucketing: allow testers to force enable/disable for themselves only
- Tagging: resulting in smaller datafiles for your application(s)
- Tracking: integration with any analytics tool
- Status site generator: for human friendly status reporting
- SDKs: for Node.js and browser environments, Kotlin/Swift planned for future
Upvote: | 48 |
Title: I've been on this site for over a decade, under different names. Generally I've found it to be one of the better tech-focused and smart online communities. However, I've recently caught myself self-censoring and avoiding sharing provocative or speculative opinions, since they immediately get downvoted (often without any explanation). It seems like there's now a culture of conformity that punishes for dissenting views. Is it (A) in my head, (B) an inevitable stage of all maturing communities, or (C) something else like the state of the world at large?
Upvote: | 66 |
Title: My friend went missing last week, and we found out that she died this weekend.<p>Someone hacked her facebook account a few months ago and she lost control of it. Now they are using it to post scam videos (deepfakes?) trying to extort money.<p>Saying she was alive, sending her $$$, etc. Facebook has been unhelpful.<p>It's really a bummer for the family right now to also deal with this, so we are trying to help.<p>Can someone help?
Upvote: | 43 |
Title: Today I fixed a couple little bugs and released a new version of my minimalist CSS framework. I built this to scratch my own itch and I've been using it for a few years as the starting point for most of my own websites.<p>I noticed I hadn't shared this with HN for about 18 months (5 or 6 minor version changes) so I thought it might be okay to do that now.<p>Edit: Err, I missed a post from 7 months ago.<p>Previous Posts:<p><a href="https://news.ycombinator.com/item?id=32990838" rel="nofollow">https://news.ycombinator.com/item?id=32990838</a>
(7 months ago)<p><a href="https://news.ycombinator.com/item?id=24594381" rel="nofollow">https://news.ycombinator.com/item?id=24594381</a> (Sept 25, 2021)<p><a href="https://news.ycombinator.com/item?id=26308052" rel="nofollow">https://news.ycombinator.com/item?id=26308052</a> (March 1, 2021)<p><a href="https://news.ycombinator.com/item?id=25660317" rel="nofollow">https://news.ycombinator.com/item?id=25660317</a> (January 20, 2021)<p><a href="https://news.ycombinator.com/item?id=23221957" rel="nofollow">https://news.ycombinator.com/item?id=23221957</a> (May 28, 2020)
Upvote: | 213 |
Title: I build ChatGPT-2D with the idea to enable people to branch AI conversations interactively, pose contextual questions based on AI response and visualize your entire dialogue on a 2 dimensional map interface. What do you think?
Upvote: | 175 |
Title: This book is considered a classic but the latest edition was published almost 20 years ago. Is it still relevant? Does it still succeed in its stated goal of bridging the gap between research and commercial software development?
Upvote: | 44 |
Title: Hi friends,<p>I started casually working on scrapscript in 2015. I built a few compilers over the years to test out various ideas/implementations, and I think I'm finally happy with the overall design.<p>The code is not public yet. Email me at [email protected] if you're interested in joining the core team later this year.<p>Let me know if you have any questions or feedback :)
Upvote: | 503 |
Title: GB Renewables Map an energy experiment created entirely in my free time (day job is visualisation at Octopus Energy).<p>It's an interactive map showing live generation for major wind farms in Great Britain, showing what each wind farm is generating both now and in the past, and where that generation is physically located.<p>Animated weather data is from WeatherLayers and shows current and historic wind conditions on the map, providing context to wind generation around the country.<p>History mode allows you to go back in time and see wind generation and weather conditions for a particular date and time. It's great for exploring days of record generation, such as the 21.6GW record on January 10th, 2023!<p>Prediction mode lets you see what wind farms are estimated to be generating using current wind conditions and model based on historic generation and wind speeds. Is a wind farm generating as you expect, or is there something to look into?<p>An experimental feature allows you to see what future wind farms could be generating today (or in the past!) if they were already built and operational. If you click the "sparkle" button on the map you'll get to see what the upcoming 3.6GW Dogger Bank wind farm is estimated to generate if it was operational today.<p>There's an "About" section on the site that goes into detail on the various public data sources and how some of the features work. I also document a lot of this on my Twitter @robhawkes if you're curious.<p>This is just the start and there are many more features to come!<p>Please let me know your comments and suggestions.
Upvote: | 245 |
Title: Hi HN! I have been working on a webapp to keep learning German and today I want to share the first version. In the past, I have used books with "Short stories" to learn the language and I always wanted more stories so I thought I could put that into an app. Short stories are nice because the learner can attempt to understand words from the context of the text and it feels like less effort than a traditional language app.<p>Additionally, the app format allows for practice questions at the end and you can use LLMs to check whether the answers are correct; so the users can check themselves.<p>So this is that! Let me know if you have any questions.<p><a href="https://webbu.app/german" rel="nofollow">https://webbu.app/german</a>
Upvote: | 180 |
Title: Microsoft released earnings yesterday and showed excellent perf on all fronts. Why do 10k layoffs then?
Upvote: | 41 |
Title: I don’t know very much about electronics.<p>I was poking around a new device and the power supply was supplying 6.18 volts to a board that appears to have 5v chips…the labels are rubbed off but one has “5v” written on it.<p>Then I looked in my box of wall warts and saw I have a 6.2 volt adapter. So I am curious why would an EE use 6.2 rather than 5 volts?<p>For context, the device is a tabla drum machine. The power supply is a bit janky. It uses line level AC. I would like to just power the board directly with a barrel plug and that seems within my ability.
Upvote: | 51 |
Title: Created some free sites to play online versions of retro MS games like Minesweeper, Solitaire and others such as Sudoku & Tic Tac Toe. I learned about Cloudflare's Pages at same time to host all the code at no cost.
Upvote: | 54 |
Title: What a great but deeply buried feature.<p>Select text like “450g” and scroll right on the cut/copy/paste popup, and enjoy the conversions!
Upvote: | 87 |
Title: Hey<p>I tend to look at 5-7 different websites, apps or notifications like 10 times a day. It's email, WhatsApp, hackernews, twitter, news, RSS feed, etc etc. It's sort of non stop. I was curious if anyone else ends up in this daily checkin hell or if you've found a way to summarise it?
Upvote: | 70 |
Title: He turned around an almost dying windows to become this Kraken that has the majority dev environment in its chokehold. That's impressive
Other candidates in my list are Frank Slootman of Snowflake and Tim Cook of Apple for how he has not dropped the ball from Steve Jobs. What are yours?
Upvote: | 67 |
Title: When you read nonfiction (i.e. Designing Data Intensive Applications, Thinking in Systems) how do you 'lock in' your understanding and apply it to your daily life?<p>Have tried note-taking, book reports, services like Readwise personally.
Upvote: | 49 |
Title: Maybe you've created your own AR program for wearables that shows the definition of a word when you highlight it IRL, or you've built a personal calendar app for your family to display on a monitor in the kitchen. Whatever it is, I'd love to hear it.
Upvote: | 1901 |
Title: Across many different pages including science journals, ChatGPT, and many others, CloudFlare verification goes into an infinite loop of:<p>1. "Verify you are a human"<p>2. Check the box or perform some other type of rain dance<p>3. "Please stand by, while we are checking your browser..."<p>4. Repeat step 1<p>I'm on Fedora Linux 37 using Firefox 110.<p>The workaround is to use Chrome.<p>After experiencing this dozens of times and getting annoyed of needing to use Chrome, I finally went and deleted all my cookies and cache which I had been dreading to do.<p>It did not help.<p>I don't have a CloudFlare account so I wrote up a detailed post on their community forums. I offered a HAR file and was willing to do diagnostics. It received no responses and it was auto-closed.<p>It's unacceptable that CloudFlare is breaking the internet while offering no community support.<p>Edit: I'm in Texas. I'm not using a VPN or Tor, just AT&T Fiber. I don't have ad-blockers. No weird extensions. Nothing special (besides being on Linux).<p>Edit2: Since this got traction, I opened a new community post: https://community.cloudflare.com/t/infinite-verify-you-are-a-human-loop/503065<p>To be clear, I'm not against CloudFlare doing DDoS protection, etc., but it can't be breaking the internet <i>while ignoring community posts on it</i>.<p>Edit3: The CloudFlare team has engaged. Thank you HN!
Upvote: | 615 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.