Show HN: Primescript – Academic Markdown editor with vector search and LaTeX https://bit.ly/3TkQCM1

Show HN: Primescript – Academic Markdown editor with vector search and LaTeX https://bit.ly/4y3vZDb September 22, 2026 at 03:29AM

Show HN: Jev Powered Obsidian Search https://bit.ly/4hlr9dB

Show HN: Jev Powered Obsidian Search https://bit.ly/3Ti8Cqj September 22, 2026 at 01:05AM

Show HN: Gdocs-me-up: a high-fidelity Google Docs exporter https://bit.ly/4hcNeL5

Show HN: Gdocs-me-up: a high-fidelity Google Docs exporter Over the years I had written dozens of Google Docs documents, that I wanted to post to my website. Not only Google Docs's publish feature is broken and incomplete, Google does not index Google Docs either. So I set out to fix it for myself using AI agents. Sharing here in case someone else finds it useful. https://bit.ly/4hgd9BQ September 21, 2026 at 01:46AM

Show HN: Try Omarchy for Windows – Full Omarchy Desktop on Windows https://bit.ly/4xwXuEl

Show HN: Try Omarchy for Windows – Full Omarchy Desktop on Windows No VM setup or configuration. Download TryOmarchy.exe, follow the prompts, and you'll be booted into a full Omarchy desktop within minutes. No dual booting or repartitioning required. Under the hood it uses QEMU + WHPX, with GPU acceleration where supported. https://bit.ly/4yIFCXW September 20, 2026 at 11:56PM

Show HN: Lightspeed, bidirectional WebSockets for Laravel (with asteroids demo) https://bit.ly/4rpbTko

Show HN: Lightspeed, bidirectional WebSockets for Laravel (with asteroids demo) Hi HN! Laravel has worked with websockets for years but generally speaking it's a one way push architecture (i.e. http in, sockets out). Lightspeed is kind of like node in that it's a single server that serves your http and your websockets all under one roof. Auth runs in 0.03ms (a Redis read, no database). Your Laravel handler runs inside the full container and the reply comes back down the same socket. I think this could be really cool for creating new ways of working with Laravel. Anyway, I put together a live demo to show it in action. Basically a cross between slither.io and Asteroids. Game here: https://bit.ly/4h63cba Repo here: https://bit.ly/4hf9RP5 Note: This is 0.1.0, the API may change before 1.0. It needs the Swoole extension and Redis. MIT. September 21, 2026 at 12:02AM

Show HN: Set, offline Markdown note-taking app with device-to-device sync https://bit.ly/4xxa0mY

Show HN: Set, offline Markdown note-taking app with device-to-device sync Hi HN. I am an avid user of Notion ( https://bit.ly/4rjCQG2 ) and really like the ergonomics of its block-based editor, but with how the product is evolving I felt I was missing a simple "single-player" experience of just taking notes and jotting down my thoughts. Hence, Set exists. It's in beta and all feedback is appreciated. You can use the app in the browser but notes are saved only to that browser session. The device-to-device sync and local dictation is desktop only. All notes are saved as markdown files in a folder you choose on the desktop app. Source: https://bit.ly/4romjRk Releases are available for macOS, Windows (x64), and Linux (x64 only for now): https://bit.ly/4779h14 https://bit.ly/4xEcQql September 19, 2026 at 11:26PM

Show HN: Flyover Country – guide to the window seat with AR https://bit.ly/4iDfhG7

Show HN: Flyover Country – guide to the window seat with AR Just released a ground-up rebuild of our long abandoned app Flyover Country. Its whole thing is to tell you fun things about what you're flying over. I'm particularly excited about the new augmented reality mode that overlays maps and points of interest on your camera view as you point your phone out the window. The 3d window view mode is also pretty neat! It's free and works offline, just load your flight up beforehand. This is a tough app to test for obvious reasons so any feedback from folks trying it out in the sky is suuuuper valuable. Please email me at shane@flyovercountry.app with any notes! https://bit.ly/4iE6JP8 September 19, 2026 at 07:13PM

Show HN: Seal – Letters and passwords that open for your family after you die https://bit.ly/4hcXXFp

Show HN: Seal – Letters and passwords that open for your family after you die I'm a CS undergrad who has developed an interest in hardware keys ever since I used them for a school project a few years ago. I made a little chat app that only allows users to chat if they registered a key, nothing else. The backstory of how I came to work on Seal (this project): was that I was generating ideas with Claude to come up with an idea that only hardware keys could make into software. I was thinking about physical house keys and what you could do with those, but in software. I was already making iOS apps with Claude, and I came up with this idea and it sounded good. I had already worked on a few projects involving hardware keys and I already loved them. Today I submitted my latest app, Seal, for review in the app store. Seal is an iOS app for iPhone and iPad that lets you write an Envelope (an encrypted bundle of passwords, photos, voice memos, etc...) for all of your loved ones when you're gone. You can register hardware keys and assign identities for each person. Then you can create Envelopes and assign them to the key that you want to distribute. They plug in the key to their phone and are ready to receive an Envelope from you. If you don't log into the app for months, one of your designated key holders can start the process of opening the Sealed Envelope. You, as the writer of the letters, get weeks of warnings to check in at least once before the letters open. After 90 days of countdown, 21 days of warnings, and 14 days of grace, the envelope opens for each person they met IRL and exchange keys with. No one else can read the envelope or see the photos (even the key holders, Apple, or me) unless that envelope was designated for that person. What's different from a password manager like 1Password: emergency access there hands one person your whole vault through the company's server, so it needs that company to still exist and be honest on the day it matters. Seal gives each person their own envelope nobody else can open, and there's no company in the middle. There's also no server at all besides a free Cloudflare static site. The encrypted bundles of passwords and photos are in Cloud Kit. Sign-in/identity is a passkey Face ID/Touch ID or a Hardware Key. It's all Crypto Kit, so there's no dependencies required. Currently, it's just me and Claude working on this. It hasn't been independently audited by anyone, and it's iPhone only for now. Please do not put a live seed phrase in it yet. If anyone has advice for me as the lead developer for this project, I'm all ears! You can help contribute on the github source code repository page if you'd like. My mom likes this app especially because she's into crypto and she needed a way to pass down or give away the seed phrases for her wallets to all 3 kids without any question about who gets what and when. In the end, there's no confusion or hurt feelings after the fact with just crypto. Thank you for reading and let me know if you have any questions or concerns for me or the project! Source: https://bit.ly/3TCrQY0 Site with limits and objections pages: https://bit.ly/4y6k7R9 Free on TestFlight while the App Store version is in review: https://apple.co/4hdUwhU https://bit.ly/3TCrQY0 September 19, 2026 at 05:28AM

Show HN: An OSS Python dependency scanner for exploited, unmaintained packages https://bit.ly/4yN5Qsp

Show HN: An OSS Python dependency scanner for exploited, unmaintained packages I built an open source python dependency scanner that will scan and flag packages with known exploit CVEs(CISA's Known Exploited list and FIRST EPSS) and unmaintained packages that have not had a release or commit in 2 years. Theres also claude hook that will make your AI agent not install these type of packages included in this repo. The full mechanism is in the readme of the project, this was just a brief summary. https://bit.ly/4yDtdoe September 19, 2026 at 01:27AM

Show HN: LiveWorld – Every 24/7 YouTube live camera on one globe https://bit.ly/4xuYUin

Show HN: LiveWorld – Every 24/7 YouTube live camera on one globe https://bit.ly/3TEUoA3 September 19, 2026 at 01:30AM

Show HN: TypeSeer – on-device autocomplete for every text field on macOS https://bit.ly/46x6374

Show HN: TypeSeer – on-device autocomplete for every text field on macOS https://bit.ly/4yGcOzd September 18, 2026 at 10:35PM

Show HN: An open Add/Search evaluation framework for agent memory https://bit.ly/4xBEbd1

Show HN: An open Add/Search evaluation framework for agent memory Hi HN, We are trying to make different Agent Memory systems comparable without letting each team choose its own answer model and evaluation pipeline. https://bit.ly/4cTYpap September 18, 2026 at 03:52AM

Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash https://bit.ly/4yIMoNl

Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash Hey HN, Henry from Cactus here. We submitted Needle 2 here a few weeks ago, and the feedback in the discussion thread was incredibly valuable, thanks! Thanks to all that feedback, we’ve been able to move quickly to release Needle 3 and I'd love to hear what you think again. The key features: 1) Automation (tool calls & structured JSON output): Needle still doesn't chat by design, its quite challenging to pack general capacity into such small models, so we focus on tool calls and structured JSON. If no tool you declared fits the request, you get an empty list back (note for when playing with the demo). 2) Intelligence Laddering: Every layer (2 to 20) is a deployable subnetwork, so one set of weights, 25 to 121 million parameters at 2-bit, shipping as 8-29MB binaries. On a Raspberry Pi 5 it decodes at up to 4k tokens/sec and prefills at up to 10k. 3) Monarch Hadamard MLP: replaces the dense FFN with three learnable Walsh-Hadamard-initialized Kronecker (Monarch) factor pairs interleaved with per-channel diagonal scales, fixed permutations, a SiLU nonlinearity, and a rank-8 input-conditioned gate, so each token gets a fully mixed nonlinear transform of its d_model channels at O(d√d) parameters and compute instead of the O(d²) a dense 4x-expansion MLP would cost. 4) Performance: On Mobile Actions (phone commands, scored on the exact call) the 20-layer model gets 86.0 through the shipped 2-bit binary; LFM2.5 1.2B is at 82.4, Qwen3.5 0.8B at 76.0, Apple's on-device model at 57.6, all at f16. More results on the link, we do not win everywhere ofc. 5) Multilingual: Needle 3 now supports English, French, Spanish, German, Dutch, Italian, Polish, with more languages coming. 6) Finetuning: You can achieve DeepSeek v4 Flash grade performance on a narrow task with just 4L, stress on "narrow task", we found that production users often prefer tuning before production. 7) Triggers: Grounding is a common challenge for tool call, at least for Needle 2, so we added support case-insensitive regular expressions matched against each request to gate false negatives. 8) Confidence: Every response also carries a calibrated confidence score, the minimum of a judgement on the finished call and its decode probability. Act above your threshold, show the call and ask below it, or escalate to a bigger model. 9) Supported Platforms: macOS, Linux on x86-64, ARM64, ARMv7, RISC-V and MIPS32, Windows x64 and ARM, Android, iOS, watchOS, tvOS, the browser as WebAssembly, and a WASI component. Thanks for reading and as always, thoughts appreciated! https://bit.ly/4yIMpAT September 18, 2026 at 01:11AM

Show HN: I made a browser tool that creates newspaper and book text match cuts https://bit.ly/4rskvqr

Show HN: I made a browser tool that creates newspaper and book text match cuts https://bit.ly/3SVqN5g September 17, 2026 at 11:47PM

Show HN: Smart Mouth Billy Bass – another fishy LLM assistant https://bit.ly/4y38jz5

Show HN: Smart Mouth Billy Bass – another fishy LLM assistant Hello, HN people. This project was inspired by another very similar Show HN post from a few months ago called BillAI Bass. I decided to do something like it but take it in a different direction. It’s not a serious assistant and doesn't turn my lights on and off, but it has no cloud dependency and was fun to build. Hope you like it. https://bit.ly/4yG30Fx September 17, 2026 at 02:49PM

Show HN: I built a new version of my fun spatial 3D online meeting app https://bit.ly/3SOZV6X

Show HN: I built a new version of my fun spatial 3D online meeting app Hi HN! flat.social is a fun spatial online meeting app for remote teams and communities. It's largely built like a web multiplayer game, and the 3D virtual spaces are highly customisable with elements and activities (I recently added surfing!). Live demo to try it out: https://bit.ly/4cQRNth Here's a quick demo video: https://www.youtube.com/watch?v=GeinBNqL23s It's a solo, bootstrapped project built with Three.js, LiveKit and Rapier for physics. I built the first version during the pandemic and posted it here back then: https://bit.ly/3xN9kNJ I had a couple of months off this year, so I locked myself in an (almost) jungle house in Brazil and rebuilt it into what I've always wanted it to be. Would love to hear what you think! https://bit.ly/3xHUoAw September 17, 2026 at 01:57PM

Show HN: Now You Can Use – Web Baseline Timeline https://bit.ly/4ygKlAx

Show HN: Now You Can Use – Web Baseline Timeline I wanted an easy way to see what was actually new AND usable on the web, something I wished caniuse had, so I made it (with AI assistance). I hope others can find some use as well! https://bit.ly/4raV3Wc September 16, 2026 at 01:54AM

Show HN: Thurbox – A tmux-based TUI and CLI for local AI agent orchestration https://bit.ly/4rcmQpg

Show HN: Thurbox – A tmux-based TUI and CLI for local AI agent orchestration https://bit.ly/4iw5uBF September 16, 2026 at 02:25AM

Show HN: A chess analyzer that runs in the browser https://bit.ly/4rdljzt

Show HN: A chess analyzer that runs in the browser I play chess casually and usually struggle to figure out why Lichess analysis say my move is a blunder. Almost all the analysis tools will just suggest the best variant for a position, and you will have to follow the suggested engine lines, or play them out manually to analyze it better. So I built this tool to make that process easier for beginner players. Behind the scene: This app run Stockfish in your browser, at a position, the app will follow the engine's suggested lines, search for any meaningful situation like a check, a fork, a capture, material change, positional disadvantages,... and point that out with an explanation. You can see the app here: https://bit.ly/4rc2H2R and paste the PGN of your game. If you play on Lichess, you can also replace the lichess.org part in the browser URl with https://bit.ly/4rc2H2R to quickly analyze, for example: https://bit.ly/4rbI8U2 https://bit.ly/4xwAXqV September 16, 2026 at 01:28AM

Show HN: Agenttik – work on multiple projects in parallel with AI agents https://bit.ly/471VQ2s

Show HN: Agenttik – work on multiple projects in parallel with AI agents Hi! I've been reading HN daily for almost 16 years. Really glad this community exist. This is my first Show HN, so I hope you find this tool useful/interesting. During my summer holidays I started playing with latest AI models on personal projects. At one point I started working in parallel with the same project and tried different approaches. The one that worked best for me was to clone the repo and distribute work among different AI workers, work in parallel but with different branches & folders. This allowed me to review and merge work afterwards. The thing is, spawning several agents in parallel was challenging. I could work on multiple features and bugfix at once, but it was hard to keep track what I was working on. I was jumping from one thing to another, and back; from one window to another. Switching tasks like this was mentally exhausting. Holidays were over, and after returning to my day to day work, I continued working using a similar approach not on one, but also on multiple projects, but I still could not find how to work in a more organized way without the mental toll. This space is growing really fast, and I haven't been able to find the right tool for myself, so I decided to build my own. I wanted a tool that would help me manage multiple projects at once, work with multiple agents, even from different providers, where I could chose between my personal and work subscriptions depending on the project, where I could easily review changes without leaving the tool if I wanted, or go all in and delegate fully to agents; one where I had quick shortcuts to move around, where I could schedule work or enqueue work for later so agents could work on one task after another. agenttik is my way of tackling this problem to work on multiple projects in parallel, while allowing myself to be on the driver seat. I can give as much autonomy as I want to the agents while keeping things organized in my head. I've been hesitating for hours on whether to submit or not, but hey "if you are not ashamed of your project, you are shipping too late" Honestly, this project has exactly one week of development, started from scratch, and even though it is still early on, it has been extremely useful for me. I've built in a week things that would have taken me months, not only built the tool itself but other projects as well (most of the merit is still on the models used, but would have had no way of distributing all the work, though). Feel free to try. Here's how I would get started: - Open a project you already have - Add some general prompt (in the project area) or add to your AGENTS.md/CLAUDE.md (to instruct it to do regular commits and/or work with branches, and also instructions on how to build effective tests so that it can self-correct). - Chose your subscriptin of choice: Claude Code / Open AI subscription / ... - Pick big models (Opus High to Max / Fable Medium to Max / Astra Medium to Max) for bigger autonomy and try not to do hand-holding Finally, start sending/enqueuing tasks on all the features you want your product to have, bugfixes, ... use a new prompt per task (Ctrl+T / Cmd+T), I recommend using big models to avoid investing your own time reviewing, and then once LLM is done, review and archive that task or continue iterating. Keep adding tasks as you think on them. Add more projects as you start feeling comfortable. It has been really addictive for me to see the speed of development when working like this. I got totally hooked. Speed of iteration is insane. Happy to discuss any feedback you have even if it is about other tools or ways of working with AI these days! I hope you like it!! https://bit.ly/3TCdua1 September 16, 2026 at 12:24AM

Show HN: Redis City – Explore how Redis works in an interactive 3D model https://bit.ly/46Xbegu

Show HN: Redis City – Explore how Redis works in an interactive 3D model https://bit.ly/4rbJOwN September 12, 2026 at 08:43PM

Show HN: Hazzel – a tiny coding agent, Bring your own keys https://bit.ly/3VaNFyc

Show HN: Hazzel – a tiny coding agent, Bring your own keys https://bit.ly/4xUDEnj September 15, 2026 at 01:15AM

Show HN: An open-source control plane for your company's AI agents https://bit.ly/4h13mPS

Show HN: An open-source control plane for your company's AI agents https://bit.ly/4h4LhAf September 15, 2026 at 12:16AM

Show HN: Macros with a Behringer FCB1010 MIDI Pedalboard in macOS https://bit.ly/3UO1lza

Show HN: Macros with a Behringer FCB1010 MIDI Pedalboard in macOS https://bit.ly/4rbeLRP September 15, 2026 at 12:01AM

Show HN: Dep-radius, know which dependency updates touch your code https://bit.ly/3UMhc1a

Show HN: Dep-radius, know which dependency updates touch your code https://bit.ly/4iZoJn9 September 13, 2026 at 11:57PM

Show HN: Surraura – audio on your Mac in a simulated environment (Helps my ADHD) https://bit.ly/46UQwxP

Show HN: Surraura – audio on your Mac in a simulated environment (Helps my ADHD) I built an app to help my late diagnosed ADHD - through sound. The app wraps whatever is playing on mac through device convolution (old radio, hifi system) - Room convolution (a bedroom, a forest) and adds ambience (rain, wind, waves) and Apples spatial processing - so it sounds like a real space. For me - hearing the sound in a space really helps more than just the dry sound, the spatial information and ambience lets me focus much more. At some point I thought- Maybe others will have the same effect? So I took the care and effort to make into a nice app with the option of downloading additional environments etc. https://bit.ly/4hq0rl5 Thank you for your time, would love to hear your thoughts. September 13, 2026 at 06:17PM

Show HN: Everything a web page can learn about you, in plain English https://bit.ly/4yAdVRc

Show HN: Everything a web page can learn about you, in plain English https://bit.ly/4yuuWfq September 12, 2026 at 11:50PM

Show HN: See Sounds on Your Webcam https://bit.ly/4illDK2

Show HN: See Sounds on Your Webcam https://bit.ly/4imABPW September 10, 2026 at 03:43PM

Show HN: We built a local-first Android agent. Then Meta launched Muse https://bit.ly/4iZAhXt

Show HN: We built a local-first Android agent. Then Meta launched Muse https://bit.ly/4xR2CEh September 12, 2026 at 10:26AM

Show HN: Don't Hit Send – the model answers while you type https://bit.ly/3UPgv7c

Show HN: Don't Hit Send – the model answers while you type Type. The model is already answering. There is no send button. Left pane is one long draft. Right pane is a stack of replies. Pause for ~350ms and it fires a normal streaming chat completion with the whole draft. Type again and it aborts the last request if that reply never produced text; if it did, that bubble stays and a new one stacks. Bubbles never rewrite. Enter is a newline. That is overlapping unary streams, not a duplex socket. Same shape as ghost-text, pointed at a conversation instead of a code line. The bit that took the work is the hold: do not fire on "and N" while someone is still typing "and NASA". git clone https://bit.ly/3USgQpE cd dont-hit-send export SCALATTICE_API_KEY=slt_... # or OPENAI_API_KEY + OPENAI_BASE_URL ./run.sh # https://bit.ly/3T0aCDG Stdlib Python, MIT, key stays on your machine. Defaults to Scalattice OpenAI-compat; any host that speaks /v1/chat/completions works from Settings or env. Browser demo on our inference platform (sign-in after a short try): https://bit.ly/4hmG88a Why've we built this? The conventional AI chat interface is overdone and lacks innovation, I've personally been building agentic software for a while now and feel a lack of innovation in the interactivity. This is a step towards trialling some different inference interfaces! https://bit.ly/4h0mnBV September 12, 2026 at 06:06AM

Show HN: Next Notes – local dictation, meeting notes, and actions app https://bit.ly/46ZTUrb

Show HN: Next Notes – local dictation, meeting notes, and actions app https://bit.ly/4xoO4ur September 12, 2026 at 03:27AM

Show HN: Graphify C# – Compiler-accurate Find Usages for coding agents https://bit.ly/4radWZu

Show HN: Graphify C# – Compiler-accurate Find Usages for coding agents https://bit.ly/3ToX2Kc September 12, 2026 at 01:16AM

Show HN: Kern Agent – See inside your agent's brain https://bit.ly/3UQ1aTQ

Show HN: Kern Agent – See inside your agent's brain https://bit.ly/4lTogTc September 11, 2026 at 02:57AM

Show HN: Learn programming by building something you want https://bit.ly/4xi5tVo

Show HN: Learn programming by building something you want https://bit.ly/3UQ8dfi September 11, 2026 at 01:20AM

Show HN: Dbmask – Discover, mask, and verify sensitive data in SQL databases https://bit.ly/4xh0vII

Show HN: Dbmask – Discover, mask, and verify sensitive data in SQL databases Hi HN I developed *dbmask*, an open-source Python tool designed to discover sensitive columns in SQL databases, masking them with deterministic fake values, and validate the masking results. The motivation behind this tool stems from a common issue I encountered when handling test data: existing masking tools often assume the user already knows exactly which columns contain sensitive information. Furthermore, the successful completion of a masking task does not guarantee that all sensitive values have actually been replaced. dbmask aims to integrate these steps into a unified workflow rather than treating them as three separate tasks. During the detection phase, the tool makes determinations by combining column names, sampled values, database type/context, historical decisions, and optionally LLMs. While LLM functionality is disabled by default, it supports local execution via Ollama or other services compatible with the OpenAI API. If the tool cannot determine a column's nature with confidence, it marks it as `UNKNOWN` rather than simply classifying uncertain cases as "non-sensitive." I also wanted to incorporate human decision-making into the system. Detection results can be exported for manual review, and confirmed decisions can be saved in CSV, XLSX, or Markdown formats, or stored in a SQL history decision table. The system logs information about both the analyst and the reviewer and re-evaluates past decisions if column types change or review results expire. The masking process employs deterministic mapping, ensuring that identical original values consistently map to the same synthetic values across different tables. Once masking is complete, the `dbmask validate --strict` command compares the masked copy with the original data, checking row counts and table structures, and verifying sensitive values row-by-row using primary keys wherever possible. The project is currently in its early stages (version 0.1.x). The current CI (Continuous Integration) pipeline covers SQLite testing; integration tests for PostgreSQL and MySQL, as well as public detection benchmarks, are included in the future development roadmap. I would particularly appreciate feedback on the detection rules, the manual review/history tracking workflows, and the verification methods. Please let me know if you encounter any usability issues or potential security risks while testing on the temporary database. Install: pip install dbmask https://bit.ly/3Vnsmtg September 10, 2026 at 04:15PM

Show HN: What if the speed of light was 5 km/h? https://bit.ly/4ilrAqo

Show HN: What if the speed of light was 5 km/h? I've always wanted to make a visualization where the speed of light was scaled down to human speeds, so that we could intuit relativistic effects with everyday objects. And here is the first version of it! https://bit.ly/4Ak5Sts September 10, 2026 at 02:58AM

Show HN: Browser-based Nginx config generator with live preview https://bit.ly/4hkqMQz

Show HN: Browser-based Nginx config generator with live preview Generates Nginx server block configs for static sites, PHP, Node.js and reverse proxy. Live preview updates as you toggle options. Download the config file directly. No signup. Runs in the browser. Part of a free sysadmin tools site — also has SSH config builder, .htaccess generator, DNS and SSL tools. Looking for feedback. https://bit.ly/4ijaglV September 9, 2026 at 01:22AM

Show HN: VolAnti – Open-source acoustic detector for fibre-optic FPV drones https://bit.ly/4cugcVu

Show HN: VolAnti – Open-source acoustic detector for fibre-optic FPV drones This project has been a month in the making, and 9 fully functional units will be sent tomorrow to an undisclosed civilian site across the Israel - Lebanon border. All of my findings are shared online to anyone on the globe who could benefit from them, or make their own units. I'm an engineering student at the University of York, originally from Israel, studying in England. I too often have been hearing about tragic cases involving civilians in war zones being chased and attacked by small FPV drones mounted with an explosive. Especially around the north of Israel, Ukraine and Russian. My intentions for this project were clear: create a device small, portable and cheap enough for civilian use, that could drastically change the outcome of a drone infiltration through early alerting. I decided to dedicate a month of my summer to go back to the university labs and to work on this as personal project. These drone attacks are not run on normal wireless drones, instead they are run on small FPV drones that are connected to the operator through a long spool of fibre optic cable. This simply means that drone detection through Radio, the current industry standard, is not applicable. My solution was to build an acoustic detector. A device that uses 4 MEMS microphones to capture the drone's propellers' emitted noise as they cut through the air, runs that data through several algorithms to enhance the SNR, reduce noise (directional, possible confusers), recognises their comb looking shape when plotted by frequency, and alert the user through multiple outputs (display, beeper, buzzer, led) along with alerting all nearby units using a LoRa module also fitted in the device. The units have been tested against a custom drone rig, housing the exact same specs as the drones used in the Israel-Lebanon border (four 2807 motors, 7 in tri blades, FPV airframe), which was stationed 104.2m away from the device in a street with passer-by's and mild wind. That test ran successfully and the device was able to detect the drone. This device was built for detection and alerting only, so no jamming. Every online resource i could think of in regarding this project exists in the Github, along with a neat website that shows off more media with explanations, also housing a live detector simulation that runs in the browser: https://bit.ly/4xQf1IC https://bit.ly/4gVfWQq September 9, 2026 at 01:08AM

Show HN: Auto-Rig Web – Zero-click 3D humanoid rigging in a Web Worker https://bit.ly/4yjK3IC

Show HN: Auto-Rig Web – Zero-click 3D humanoid rigging in a Web Worker https://bit.ly/4yrPpBp September 8, 2026 at 11:48PM

Show HN: Zero downtime embedding model upgrades https://bit.ly/4ctEEq2

Show HN: Zero downtime embedding model upgrades People use embedding models all the time for rag/semantic retrieval. However, when a newer, more desireable model comes out, there is an expensive (both in time and computational) cost of re-embedding every document in the database. However, I figured out an interesting way to forgo that upfront embedding cost. algo: old model/index -> retrieve top-K docs -> score those docs with the new model -> cache/materialize the new embeddings so instead of rebuilding the entire vector store upfront, the old index keeps getting retrieved from, while the new model reranks those candidates. This works surprisingly well for some model pairs, (i tested 63 source-> target migrations on h100s, on upto 1M documents). For example, on a 1M document Natural Questions dataset, native Qwen3-Embedding-8B: 0.6812 nDCG@10 Qwen3-4B -> Qwen3-8B, K=50: 0.6816 Qwen3-0.6B -> Qwen3-8B, K=50: 0.6638 MiniLM -> Qwen3-8B, K=50: 0.6486 (the hard part is determining k, I held the k constant above to give some sense of migratability). You can install it with pip pip install embedflow and the code is on github https://bit.ly/4zXvTyx https://bit.ly/4zXvTyx September 8, 2026 at 03:35AM

Show HN: Browse 27 years of movie ticket stubs in ThreeJS https://bit.ly/4gVMLNe

Show HN: Browse 27 years of movie ticket stubs in ThreeJS This is one of those projects that's been sitting in the back of my mind for years. I made a ThreeJS UI for browsing (read: re-living) the ticket stubs I've been collecting since I was 9 years old. I was going to create a whole form so that people could share their own collections but landed on just opening up a version you can use to DIY with your agent. That part is here: https://bit.ly/46JQf0y Enjoy! Notes and feedback welcome and appreciated. September 8, 2026 at 12:32AM

Show HN: Golden hour API and a 3D globe of the best light https://bit.ly/4r1NiC3

Show HN: Golden hour API and a 3D globe of the best light https://bit.ly/4r3TERy September 7, 2026 at 05:18AM

Show HN: Lantunnel – a P2P-first private mesh for reaching your LANs https://bit.ly/4cw9s9u

Show HN: Lantunnel – a P2P-first private mesh for reaching your LANs https://bit.ly/4gMZ5iU September 7, 2026 at 04:47AM

Show HN: GET Together – A social network where you don't need POST to Post https://bit.ly/3SAswNb

Show HN: GET Together – A social network where you don't need POST to Post https://bit.ly/4zUnrjD September 7, 2026 at 02:41AM

Show HN: Keen Bean – Mac meeting notes that draft the spec while you talk https://bit.ly/4iaEF5N

Show HN: Keen Bean – Mac meeting notes that draft the spec while you talk https://bit.ly/4gMY769 September 6, 2026 at 10:53AM

Show HN: OpenMonitor – vendor neutral web monitoring cloud agent https://bit.ly/4r2ShCE

Show HN: OpenMonitor – vendor neutral web monitoring cloud agent https://bit.ly/4hbAPIC September 6, 2026 at 01:03AM

Show HN: Cowbull – Wordle-like game where you only get match counts https://bit.ly/4A4lscC

Show HN: Cowbull – Wordle-like game where you only get match counts CowBull is a word game where you have to find the hidden four-letter word. You make guesses and get back the count of exact letter matches (bulls) and partial matches (cows). It’s a bit like Wordle, but you are not told the exact letters which match, only the count of matching letters. Scoring is based on the number of guesses and time used. There is a single player mode and a multi-player mode (no account required). The game source code is at https://bit.ly/4gQELgB . When I was younger, we used to play this game on paper at the back of the classroom during school and college in India. During Covid, I made this online version of the game to play with my kids. https://bit.ly/4xMYAwI September 6, 2026 at 01:37AM

Show HN: I made Blocknado, an orbital stacker. Think Tetris with square rings https://bit.ly/4ym4lRC

Show HN: I made Blocknado, an orbital stacker. Think Tetris with square rings I've loved stacker type games for as long as I can remember. But I've always wanted a version with a bit more strategy and puzzle dynamics. Blocknado is a game I've always wanted to play, maybe others will too. https://bit.ly/3SzxUjE September 6, 2026 at 01:16AM

Show HN: TimerVana a Local-first scheduler for real world routines https://bit.ly/4cNMJ9d

Show HN: TimerVana a Local-first scheduler for real world routines Hi HN, I’m the creator of TimerVana. It's not just "another timer app". The engineering analogy is a lightweight task scheduler for things people do in the real world. A normal timer models one countdown. TimerVana models a small dependency graph in JSON: some steps run sequentially, others run in parallel, and everything can converge on a target finish time. You can also combine multiple timers (e.g., combine a timer with all the steps to cook a steak + a timer for roasting vegetables). I originally built it because I often set multiple timers and alarms when cooking and got tired of repeatedly juggling separate alarms to get a main dish and several sides ready together. The same model also works for workouts, study sessions, morning routines, and other repeatable processes. The iPhone app is local-first by design [0]. You can create, save, and run timers entirely on your phone without an account or cloud sharing. If you want the networked features, the companion website lets you browse and download timers, upload your own, and share them with other people. Everything available today is free. I may eventually add optional AI-assisted timer creation as a paid feature, but I haven’t settled on the business model or pricing. I'm more doing this to solve my own problem and hope it will help others as well. I’m interested in whether the underlying abstraction makes sense: does a sequence of dependent and parallel timed steps help you? I’d also appreciate feedback on the balance I took between the completely local experience and the optional sharing layer (I was partially inspired by the old iOS Workflow App, which was acquired by Apple and is now called Shortcuts). Of course, any other feedback very much appreciated. [0]: https://apple.co/4zXnb3g... https://bit.ly/4xcAWbQ September 4, 2026 at 11:47PM

Show HN: Interactive map of the Japanese address system https://bit.ly/4xbIziG

Show HN: Interactive map of the Japanese address system I had a hard time understanding the Japanese address system. Parts of addresses have suffixes like fu, shi, ku, chōme, ban and gō. Which suffixes are used varies between regions in a way that seemed random to me. I couldn't find a map of the system with a modern visualization, so I made one. It also shows whether an area has a governor or a mayor. For example Tokyo's wards have mayors and Osaka's don't. You can click around or drop the pin to explore, or write an address in the search bar. https://bit.ly/4iLta4P September 4, 2026 at 07:42AM

Show HN: Word42, a wordprocessor inspired by MS Word and Abiword https://bit.ly/461SZGu

Show HN: Word42, a wordprocessor inspired by MS Word and Abiword https://bit.ly/461I7IH September 4, 2026 at 05:39AM

Show HN: Reactor Atlas https://bit.ly/3Vc6aSN

Show HN: Reactor Atlas Hi HN, I'm a nuclear engineer and tech entrepreneur. After working for 3 years in the National Atomic Energy Commision (CNEA) in HPC software for nuclear simulations, I co-founded ratherlabs.com, a software company providing blockchain solutions for startups. Rather Labs business reached ~6 million usd in annual revenue in 2024; now demand is decreasing as our niche clients are using AI to launch their MVPs. Now I'm joining my scientific background with my entrepreneurial inclinations to think of problems related to the energy industry. Nuclear is experiencing a renaissance now (it was in a dark age when I was studying after the Fukushima event) so I think this is the right moment for me to jump again into the field. Reactor atlas is an interactive nuclear reactor map and an intelligence layer on top of it. The basic layer indexes: - power reactors - research reactors - fuel facilities with their history and projection, by country. The project is solving a fragmented information ecosystem, where operational and commercial data is hard to conciliate. Also, I'm monitoring nuclear news associated with each of the facilities, political declarations that may affect the nuclear landscape and geo-phisical news (like earthquakes) that could affect operation. On top of the operational + news layer, I'm creating an intelligence layer that will help institutions of different types make faster decisions by subscribing to very specific alerts. As an engineer, I used to code in fortran and c++ when I was working at CNEA, and later I participated in the first Rather Labs projects architecting complex async systems, mostly using MERN stack. I'm no longer coding anymore. I'm focused on designing robust systems with AI agents. I mainly used Claude Fable 5.1 to develop reactoratlas.com, deploying on vercel and using mostly nextjs, threejs, postgreSQL and some cron jobs to query and update data. I miss coding as in the old days, but I haven't code a single line in this project. And I strongly believe we humans shouldn't code anymore: our responsibility is now on designing and supervising AI systems. Thanks for your attention! https://bit.ly/3US1Vf6 September 3, 2026 at 01:33PM

Show HN: I made a word building game supporting anagrams and one handed use https://bit.ly/4yk2BIL

Show HN: I made a word building game supporting anagrams and one handed use Hi everyone, Wanted to share a game I built. It's called Unscramble! and I believe it's the simplest word building game. Rules: You get a few letters on screen. You need to make one word with all the letters. Tap a letter to select. Tap again to deselect. Faster you go, better you score. If you enter anagrams of the word it's looking for, it gives you extra points. If you are stuck, you can either "Ask Friend" or get Hint. You get extra points for getting the word fast but you don't get penalized for taking too long. I believe the simplicity of the game along with big touch buttons makes it perfect for one handed use. There's 6 difficulty levels in the game, and settings for things like toggling the timer, relaxing background music, sound effects, haptic feedback, removing the background animation. If you want to try: https://apple.co/4yiPgQP... Would love to hear feedback to improve it. September 3, 2026 at 03:37PM

Show HN: The cheapest GPU cloud – H100s at $2.04/HR, H200s at $3/HR https://bit.ly/4qOPIDS

Show HN: The cheapest GPU cloud – H100s at $2.04/HR, H200s at $3/HR https://bit.ly/3TcRgLw September 3, 2026 at 03:50AM

Show HN: Redactle LLM Leaderboard https://bit.ly/4qSxpO3

Show HN: Redactle LLM Leaderboard I got nerd sniped and ran a bunch of evaluations to see how well models could play Redactle. It turns out Gemini flash is in a league of it's own in both performance and cost. It's able to one-shot every eval in a just a few seconds. https://bit.ly/4zXqOq0 September 3, 2026 at 01:11AM

Show HN: Open source ML programming language playground https://bit.ly/4dk3Tez

Show HN: Open source ML programming language playground After joining a ML Study group and trying Google Colab I decided to create my own array programming language to help me learn and visualize machine learning. There are many demo repos using this language which has become more general purpose. Under active development. See github link in footer. Built in Tour, Demos, Tutorials, Glossary, etc. Can use this same UI on a local system to program Apple Silicon, NVIDIA GPUs, or to use Ollama models for context aware help in the REPL. The system architecture is documented in the wiki: https://bit.ly/4xbfGDb https://bit.ly/4yckKIf September 2, 2026 at 11:05PM

Show HN: Markdown Gatekeeper – one current source per topic for AI agents https://bit.ly/46A48hN

Show HN: Markdown Gatekeeper – one current source per topic for AI agents https://bit.ly/3V8pmAQ September 2, 2026 at 03:48AM

Show HN: Gopdfrab – Open-source PDF/A processing for Go https://bit.ly/4gABnpO

Show HN: Gopdfrab – Open-source PDF/A processing for Go https://bit.ly/4h8jDCP September 1, 2026 at 07:34PM

Show HN: Three Pins a daily 3-hole golf game in the browser https://bit.ly/4cWgTH7

Show HN: Three Pins a daily 3-hole golf game in the browser https://bit.ly/4cT7zDU September 1, 2026 at 06:45PM

Show HN: Notes Plus – A Markdown Apple Notes Clone with Latex, Excalidraw etc. https://bit.ly/4xxc8Mm

Show HN: Notes Plus – A Markdown Apple Notes Clone with Latex, Excalidraw etc. tldr: clone of Apple Notes, but every note is a local Markdown file. Hi, I used a summer of unemployment to actually finalize a hobby project, a literal clone of Apple's Notes, with open access to the filesystem, meaning each note is just a markdown file. Then implemented renderers for Excalidraw, Latex, AsciiMath, mermaid diagrams, PlantUML, Graphviz, Vega Lite, mhchem and footnotes (a feature i need a lot in texts). It can also show notes in vertical lists, like swimlanes, I call it "JIRA view" ;) Free for now (might monetize later if successful, with $19 one-time payment, NOT subscription)! I'm not 100% sure about the legal aspect of this, but i will try to even list it in the Appstore. The binaries are signed / notarized by Apple. *PLEASE NOTE:* This is an early alpha! SO BACKUP YOUR STUFF OR JUST WORK WITH COPIES! There are issues! I'm posting this here to collect hurtful user feedback. There's a bug report button in the app, and it will notify you about available updates. Thanks, I kindly request any form of feedback! https://bit.ly/4h4VBJR September 2, 2026 at 12:17AM