distribut - Moozonian Search

news.ycombinator.com• Mar 21, 2025• 1 min read

AI published a bogus CVE for my projectI help maintain Dask, an OSS project in the Python data space. Dask helps users run arbitrary Python code on distributed clusters of computers.Yesterday, I and several users got a CVE alert (https://www.cve.org/CVERecord?id=CVE-2024-10096) that there was an exploit allowing users with access to a Dask scheduler to run arbitrary code on that server using a backdoor with pickle.Obviously, the creators of this CVE missed the easily accessible `client.run_on_scheduler(my_function)` API, which makes this much easier for a would-be hacker. Is this a backdoor? Maybe, but the front door is wide open inviting all to enter.I did some research into the creator of the CVE, a project call "Protect AI" (Dask is often used in AI workloads). It looks like they acquired an AI bug bounty company, Huntr (https://protectai.com/newsroom/protect-ai-acquires-huntr), which reported the "bug" several months ago. Even though this was spam, Dask maintainers responded to the bug report saying ...> This is the exp

bing.com• Mar 5, 2025• 1 min read

250 CPR Anytime Training Kits Donated to Local HospitalsThe American Heart Association, devoted to a world of healthier lives for all, is proud to announce a donation of 250 CPR Anytime Training Kits. These kits were be distributed to Joe DiMaggio Children ...

github.com• Feb 27, 2025• 1 min read

Show HN: Botwell – A Framework for LLM Comparative Analysis Using AI Peer ReviewBotwell is an automated framework that helps evaluate LLM capabilities through a unique approach: models grade each other's responses to challenging prompts.The concept is based on this post: https://substack.com/inbox/post/157571824 where models act as both writers and critics.Here's what makes it different from traditional benchmarks:1. Peer Evaluation: Models write essays on complex topics, then grade each other's work 2. Cross-Domain Analysis: Tests across multiple domains (described below) 3. Grading Bias Detection: Measures which models grade more strictly/leniently vs. the consensus 4. Comprehensive Boswell Quotient: A 0-100 score that combines performance, evaluation capability, and efficiencyWe've built test domains across three categories:POLITICAL SCIENCE: - Level 1: AI policy analysis - Level 2: Complex AI governance with rigorous grading criteriaCOMPUTER SCIENCE: - Level 1: Algorithm analysis and complexity theory - Level 2: Distributed system design challengesPROGRAMMING

news.ycombinator.com• Feb 26, 2025• 1 min read

Ask HN: How do I finish this lottery scheme?This is a cryptography problem I haven't solved yet.[1] Maybe there is some background research in distributed computing that I don't know about yet.The State of Utopia is a distributed, AI-run, Utopian state, and has certain scarce resources such as citizenships and money that in the interest of fairness must sometimes be granted by lottery, since there is no such thing as half a citizenship for example, or it might have a Universal Basic Income level that it sets in an experiment and only has money to give to a certain number of people, and it doesn't scale to trying to run the experiment on everyone at once.However, absent trust, there is a problem with attackers who want to compromise the entire system to choose their own lottery winner - for example so they can set up a corrupt black market where they choose the lottery winners and require payment for their choice. Or they can just choose themselves as winners.How could the population of Utopia trust the random number which repres

news.ycombinator.com• Feb 12, 2025• 1 min read

DietPi released a new version 9.10DietPi is a lightweight Debian based Linux distribution for SBCs and server systems, with the option to install desktop environments, too. It ships as minimal image but allows to install complete and ready-to-use software stacks with a set of console based shell dialogs and scripts.The source code is hosted on GitHub: https://github.com/MichaIng/DietPi The main website can be found at: https://dietpi.com/ Wikipedia: https://de.wikipedia.org/wiki/DietPiThe project released the new version DietPi v9.10 on February 9th, 2025.The highlights of this version are: RISC-V (StarFive VisionFive 2, PINE64 Star64): Switch to Debian Trixie and support of Bazarr, Raspotify, NZBGet, MicroK8s and AdGuard Home Raspberry Pi, NanoPi M6: New tool DietPi-Display supports setting of console display modes/rotation Raspberry Pi: Migration to the new Raspberry Pi kernel/firmware stack is now possible via dietpi-config DietPi-Automation: New option in dietpi.txt for automated APT-based program installs myMPD: A

news.ycombinator.com• Jan 13, 2025• 1 min read

Ask HN: Is maintaining a personal blog still worth it?Remember when maintaining a blog was THE way to build your developer brand?When thoughtful technical writing could lead to speaking gigs, job offers, and meaningful connections?But in 2025, the landscape has shifted dramatically:- LinkedIn's algorithmic feed heavily favors short-form "broetry" over substantive technical content - Twitter/X has become a battleground of AI-generated hot takes - Medium is drowning in SEO-optimized tutorials that all say the same thingUnless you're already established or willing to play the AI-SEO game, it feels impossible to build genuine readership for a technical blog in 2025.Yet part of me wonders if I'm just being cynical. Maybe there's still value in writing for its own sake? Or perhaps there are distribution channels I haven't considered?For those still maintaining personal blogs: How do you find readers? Where do you share your content? And most importantly - why do you keep writing?

ismyaidown.com• Dec 24, 2024• 1 min read

Show HN: IsMyAIDown – AI status monitor with Phoenix LiveView. Can you crash it?I recently built IsMyAIDown.com to challenge some assumptions about modern web development. Instead of reaching for React like everyone else seems to be doing these days, I wanted to see if I could build a distributed, scalable, real-time, single-page app using server-side tech with Phoenix and LiveView.The core idea was simple enough - monitor major AI providers and show their status across different geographic regions. But the primary goal was to better understand what was possible without the complexity of modern frontend frameworks.I ended up deploying across three Fly.io regions and using Oban (a very good job processing library in the Elixir ecosystem) to handle geographically distributed health checks. Propogating new checks and detecting and tracking user presence was all quite straightforward. Phoenix is a great framework.That said, the mental model with LiveView is... a lot. You've got to keep track of a variety of message callbacks that all seem to run together in my head (m

retirementodds.com• Dec 15, 2024• 1 min read

Show HN: Retirementodds.com – a versatile free retirement Monte Carlo calculatorI built a retirement calculator because I couldn't find one for free that did everything I wanted. This originated as a spreadsheet that helped me with my FIRE journey, and now its my retirement passion project.RetirementOdds.com has a simple mode for getting started, but a much more capable advanced mode. In advanced mode, you can have an unlimited number of assets and cash flows. It handles specialty cash flows like investment real estate, health care, social security, and loans. Every cash flow can have custom start and end dates and rates of return or inflation. Federal and state tax expenses are estimated for you, including handling of long term capital gains.The computation is a Monte Carlo "chance of success" based on a normalized distribution of returns with an average and standard deviation that you specify. Plus eight different useful charts in the results pane. You can also do a "What If" calculation which allows you to choose one parameter to vary while it runs the calculat

github.com• Dec 1, 2024• 1 min read

Show HN: Connect any Open Data to Claude (and soon, any other LLMs)Problem: LLMs need context data to give accurate answers and thousands of great publicly available datasets remain vastly unexploited. Anthropic’s MCP servers are awesome, but we need many more servers and an easy way to discover them.What if it took only one command to connect any LLM to any open dataset?Open Data MCP, a 2-in-1 solution:1. Access Open Data: - 10 seconds setup to query any integrated Open Data MCP server from Claude (and more to come) - Simple CLI tool2. Access Open Data: - Create your own Open Data MCP server(s) - Get templates, guides, and community - Get instant distribution — your data becomes discoverable by all users via our CLIWhat’s next?A) Build many Open Data MCP servers to make all the Open Data truly accessibleB) Make Open Data available to all LLM applications (beyond Claude)The bottleneck is A)!Call to all Hackers! We need the community’s help to build MCP servers for the millions of public datasets out there, e.g., from governments, public entities, comp

news.ycombinator.com• Nov 15, 2024• 1 min read

How to convince my parents that masters in IT in Nepal is not worth the hassle?I am from south asian country.Parents are forcing me to do masters in data science and analytics. I want to do masters, but not in my home nation. And I don't really have money for doing it in Europe/US. My best bet has been on getting Assistantships and merit-based scholarships per semester. However, I do realize it's extremely competitive and luck is a huge factor as I've some friends who got C grades in engineering getting RA in the USA.I've a bachelors degree in computer science. And I must say it left a lot to be desired for me in terms of both knowledge and grades.Pros of masters degree:- Get good grades and apply for abroad for scholarship.(But extremely impossible to get good grades with full time job).- Good connections in data field as there will be lots of eager to learn students.- I've been self learning stuffs like data structures, algorithms, database, programming, distributed system etc. This degree will provide me with a structured approach to learn all these.Cons of de

bing.com• Nov 7, 2024• 1 min read

Alaris Equity: High Yielder Delivers Once MoreAlaris Equity Partners Income has been a strong performer for us. Q3-2024 results were once again strong and the stock charged ahead after the press release. The large distribution remains comfortably ...

bing.com• Oct 29, 2024• 1 min read

Accelerate root cause analysis with OpenTelemetry and AI assistantsIn today’s rapidly evolving digital landscape, the complexity of distributed systems and microservices architectures has reached unprecedented levels. As organizations strive to maintain visibility ...

bing.com• Oct 19, 2024• 1 min read

Costco Has a Magazine and It’s ThrivingEach month, 15.4 million copies of Costco Connection are mailed out to members. Another 300,000 are distributed via Costco warehouses. It is now the nation’s third largest magazine. The cover of ...

news.ycombinator.com• Oct 16, 2024• 1 min read

Fundraise news is one of the best PR tools to go viralThe way ads are created and distributed has constantly evolved. Gone are the days when actors simply held a bottle of shampoo and endorsed it, now yield negligible ROI compared to content creators claiming they use the product every day. This shift is deeply connected to how our brains have adapted, increasingly filtering out familiar ads. One key reason for this is our decreasing attention spans and the constant desire to see something new.What works?Something which is very distant from reality wrapped in a layer of creativity, works exceptionally well. For example, Johnny Sins being portrayed as impotent or Rahul Dravid acting angry—these unexpected, out-of-character depictions capture attention.However, sharing credit card details on Twitter may not work again!During our research on Product-Led Growth (PLG), we explored how to replicate this attention-capturing phenomenon for ourselves. We spoke to multiple founders and people who have successfully navigated this space. A surprising

bing.com• Oct 2, 2024• 1 min read

Que retenir de la nouvelle recommandation concernant DDA ?La première publication de l'ACPR du 17 juillet 2023 était éclairante sur deux volets majeurs de la réglementation sur la distribution d'assurance : le POG (gouvernance produit) et la délicate ...

github.com• Sep 27, 2024• 1 min read

Show HN: A tool that turns everyday computers into your own AI cloudI have a favour to ask.I’ve been working for a while on Kalavai, a project to make distributed AI easy. There are brilliant tools out there to help AI hobbyists and devs on the software layer (shout out to vLLM and llamacpp amongst many others!) but it’s a jungle out there when it comes to procuring and managing the necessary hardware resources and orchestrating them. This has always led me to compromise on the size of the models I end up using (quantized versions, smaller models) to save cost or to play within the limits of my rig.Today I am happy to share the first public version of our Kalavai client (totally free, forever), a CLI that helps you build an AI cluster from your everyday devices. Our first use case is distributed LLM deployment, and we hope to expand this with the help of the community.Now, the favour!I’d love for people interested in AI at scale (bigger than a single machine) to give it a go and provide honest feedback.Do you share our motivation?If you tried Kalavai,

news.ycombinator.com• Sep 24, 2024• 1 min read

Ask HN: Why would a CA revoke a cert with a public private key?To get working HTTPS on localhost, something I have done in the past is: - register myproject.dev, - point it to 127.0.0.1, - create a cert for it, and - just store the private key in the repo. Every coworker can check out the (private) repo and has working HTTPS without any fuss or configuration.There are projects like https://lcl.host, but they require installing stuff on the machine and/or modifying the browser trust configuration.Why has nobody just registered a similar domain like lcl.host, pointed it to 127.0.0.1, and published the private key for everyone to use?Would the CA revoke this cert? Why? Doesn't the domain owner get to define the set of servers they allow to use the cert, and if that set just happens to be everyone, so what?Is this "there are limits to how wide you can distribute your private key" policy documented somewhere?Looking at digicert[1], if a revocation request is submitted, the owner must approve it. What happens if I just don't approve it?[1]: https://docs

news.ycombinator.com• Sep 22, 2024• 1 min read

Ask HN: Has anyone been on GCP's springboard programme?GCP is inviting our startup to enroll on their springboard programme. I hadn't heard of it before and can't find anything online about it. From what they describe it's a ~3 month programme to get your product onto the GCP marketplace, which you must make a (soft) committment to do. They send some form of "integration expert" to help you get billing setup and by the sounds of it do a form of audit to make sure your product is suitable. They said this was very light touch but I'm not sure how they can say that without knowing any technical details of our product.I like the idea of opening up this new distribution channel and can see it help in procuring enterprise customers who are more likely to trust a Google certified 3rd-party. However, I'm wary of it being a big distraction. Keen to hear if anyone has direct experience with springboard or even just general advice on the process of getting onto GCP marketplace. Thank you.

news.ycombinator.com• Aug 23, 2024• 1 min read

Is that the central shortcoming in binary floating-point math?Hi, I'm struggeling with bin-FP imprecision ( think not the first one who does ), and think to have identified one central weakness.IEEE 754 standard delivers consistent arithmetic for a wide range of 'english plumber metrics' ( imperial system of measurement, halves, quarters, eights, sixteenths and so on ). Up to where it runs out of bits and switches into rounding.Alas decimal fractions are encoded in a way that most of them fall in that weak category, and operation results often need additional fit-in rounding.Rounded results aren't qualified for associativity and distributivity, the math with them is 'broken', is not qualified arithmetic from mathematical POV.In consequence the problems for calculations with decimal fractions are not limited to extreme cases or peripheral areas, but occur freely scattered even in the simplest calculations. ( 2 times 64 bit fail to add 0.2 and 0.1 correctly, embarrassing )Alas plenty of programs / users try substituting decimal fraction math with b

news.ycombinator.com• Aug 21, 2024• 1 min read

Launch HN: Outerport (YC S24) – Instant hot-swapping for AI model weightsHi HN! We’re Towaki and Allen, and we’re building Outerport (https://outerport.com), a distribution network for AI model weights that enables ‘hot-swapping’ of AI models to save on GPU costs.‘Hot-swapping’ lets you serve different models on the same GPU machine with only ~2 second swap times (~150x faster than baseline). You can see this in action through a live demo where you try the same prompts on different open source large language models at https://hotswap.outerport.com and see the docs here https://docs.outerport.com.Running AI models on the cloud is expensive. Outerport came from our own experience working on AI services ourselves and struggling with the cost.Cloud GPUs are charged by the amount of time used. A long start-up time (from loading models into GPU memory) means that to serve requests quickly, we need to acquire extra GPUs with models pre-loaded for spare capacity (i.e. ‘overprovision’). The time spent on loading models also adds to the cost. Both lead to inefficient