Ideas Lab
Real problems people complain about online, pulled every morning and scored out of 100. Build, validate, or skip. How scoring works
- Ideas
- 4,820
- BUILD
- 702
- Sources
- 8
- Last sync
- 6h ago
Open-source model routing for coding agents at Astra-level performance
A few months ago we started building a model router for coding agents because we thought we could outperform any single model with an ensemble approach. Recently we’ve achieved that milestone and I want to talk about how we did it.First of all, a quick explanation: the Weave Router (https://github.com/weave-os/router) plugs into any coding agent (e.g. Claude Code or Codex) and intelligently switches between LLMs. So, for example, Astra handles tricky debugging or complex system design tasks, and Deepseek v4 Flash handles simple frontend updates.What we’re announcing today is our new routing model, which we’re calling Weave Router 2.0. We benchmarked 2.0 against GPT-6 Astra on Terminal Bench 4.0 and SWE Atlas. On both benchmarks, the router had equivalent pass rates. On Terminal Bench, the router hit 52% of Astra’s cost, and completed tasks 2.2x faster. On SWE Atlas, the router cost 54% as much as Astra and ran 2.5x faster. (Full results on our website at https://weaveos.com/router!)It turns out training a model to route effectively - taking into consideration model capabilities, costs, cache awareness, and more - is a really hard problem! I want to talk about three ways we were able to improve so much over the last few months: 1) a new architecture, 2) larger training data set size, and 3) smarter cache-eviction impact calculation.1) a new architecture. Our initial approach used an RL model without many priors. While RL is still an important part of the story, the cost of fully exploring the space of routing decisions is very high, so we’ve taken some shortcuts that have significantly improved performance.Consider how large the search space for the routing problem is. Take a typical coding agent session, with ~100 agent turns (i.e. 100 LLM API calls). Technically there are 100 chances to select a model. If we assume a roster of ~10 models (of course there are lots more but we can remove any that are Pareto dominated), then there a
Hacker News1mo agoToolAI
Lathoa, a math app for kids where the AI is wrong on purpose
I made this for kids around 10 to 14. A robot called Errol solves a math problem step by step and one of the steps is wrong. The kid has to find it and say what's wrong with it. Sometimes nothing is wrong, so just saying "there's a mistake" every time doesn't work. The user needs to enter an explanation if she finds an error to gain more XP; speed matters also for more points. There is no direct interaction or chatting with an LLM. Lathoa's harness is stable and has many evaluation steps to catch inconsistencies and prompt injections.You can play one on the homepage without signing up.The part that surprised me: it's hard to get an LLM to be wrong on purpose. Half the time it gives you the right answer and calls it wrong, or a "mistake" that's actually correct. So every case gets checked before a kid sees it. Where it can, a plain arithmetic check redoes the math exactly. A second model also solves the problem without seeing Errol's work. If anything disagrees, the case is thrown away.The weak spot is that the second model can make the same mistake as the first. The arithmetic check is there for that, but it only works on English cases so far. German and Greek write decimals with a comma and I haven't got the parsing right yet.What I'd really like to know: does finding someone else's mistake teach anything that solving the problem yourself doesn't? I'm not sure, and I'd like to hear from people who teach.
Hacker News1mo agoToolAI
Using 2D DFT, dithering, etc. to maximize eInk manga image quality
Kindle Comic Converter optimizes black & white (or color) comics and manga for E-ink ereaders like Kindle, Kobo, ReMarkable, and more. Pages display in fullscreen without margins, with proper fixed layout support. Output works great in KOReader.Full details on what KCC does is in the readme with plenty of photos!KCC has been under continuous development since 2012 and has code sign on both Windows and macOS. Also available on Linux.-the current kcc dev
Hacker News1mo agoToolAI
AI·rete·RAG – a Rete rule engine decides, RAG explains why
Hi HN, I built ai·rete·rag because I kept seeing teams put an LLM in charge of decisions that need to be auditable (lending, fraud, clinical triage), then bolt on "guardrails" after the fact.It runs the two in series instead:1. A pure-Python Rete engine evaluates YAML rules against your facts. The verdict comes only from here. Same facts, same verdict, every time, with salience-based conflict resolution. 2. RAG retrieves passages from your own policy documents, and an LLM writes a plain-English explanation of the decision that was already made, citing those passages. It can't change the verdict.A few things that went further than I expected: - Rules are a graph, not flat lists: nested all/any/not, and rules can assert facts that other rules consume (forward chaining). The decision trace shows the causal chain. - Audit mode records every rule evaluated, including the ones that didn't fire, condition by condition, with a snapshot of the rule set for replay. - Rules can steer retrieval (a fired rule narrows which documents get searched), and retrieved text can be turned into facts for the engine. - Non-technical authors can build rules in a visual editor, or paste a policy document and get LLM-drafted rules with citations. Drafts are never saved without review. YAML is still there for engineers.The landing page has a live demo with no signup (8 demo domains: loan, fraud, clinical, insurance, legal, ops, e-commerce, blockchain). There's also an MCP server, so Claude and other agents can call /decide as a tool: `uvx ai-rete-rag-mcp`.To be upfront: it's a hosted product with a free tier. The MCP client is open source (MIT, github.com/zaharajabeen13-create/ai-rete-rag-mcp); the engine and platform are not open source right now.I'd especially like to hear from anyone who has had to explain an automated decision to a regulator or an auditor: what did they actually ask for?
Hacker News1mo agoToolAI
JBR-001 – An open-source 3D printable desktop robot
We built a desktop companion robot that is powered by Arduino UNO Q. Project is open source: source code, 3D printable files and assembly instructions are available at Arduino Project Hub.Robot is equipped with camera, distance sensor, buzzer, moves head and arms with 3 servo motors and has an animated display. It can recognise objects and respond to it. We build it with idea this should be fun project people can 3D print and build at home themself.
Hacker News1mo agoToolAI
Magnitude (YC S25) – Self-optimizing inference engine for agents
Hey HN, Anders and Tom here. We're building Magnitude, an inference engine for agents that optimizes itself to run as fast as possible on your hardware. It works on Mac, Linux, and Windows on any hardware and is up to 2x faster than llama.cpp.We're both software engineers and previously built an open source browser agent to 4k+ GH stars and 100k+ downloads. We increasingly wanted to run it on local models, but found that no inference engine worked for our use case.Inference engines today all make a performance tradeoff. They are either:- Built for batched inference on datacenter hardware at the cost of single-session performance (vLLM, SGLang) - Designed for broad compatibility instead of optimizing for specific hardware (llama.cpp, Ollama) - Specialized for specific hardware or models but lacking engine completeness (oMLX, ds4)Plus none of them are designed for running agents locally. Sessions are long, several often run at once, and you still want to use your computer for other things.Magnitude is built for maximum performance on your hardware and running local agents:- On-device compilation and tuning: Kernels are written with flexible parameters that are tuned on your actual device before the model runs. This gives you broad hardware compatibility with the same performance ceiling as hardware-specific kernels.- Focus on best architectures: We write our tunable, highly efficient kernels for the most popular open-weights families. This allows us to achieve and surpass the performance of hardware or model specialized engines, without forcing ourselves to over-generalize at the cost of performance.- Dynamic memory allocation: Magnitude reserves only enough memory up front to hold model weights. As your agent sessions grow, the memory heap dynamically increases, and frees itself when agents stop. Your hardware can still be used for other stuff while agents run.- Hybrid paged attention: We borrow the best ideas from engines like SGLang to allow concurrent se
Hacker News1mo agoToolAI
Parrot – Open-Source Smart Meeting Recorder with Co-Pilot on Mac
Hello Hacker NewsI am Uygar, I am an entrepreneur from London. I run my businesses on calls everyday, customers, partners and suppliers. And i have a horrible memory i forgot things. So months ago with the help of Claude i have built Parrot. I have designed and coded for my work and what i need. An to be honest worked very well and helped me alot.Parrot records what your mac hears and everything stays in your computer if you use local models, no account, no API calls. There is something called copilot which made my life easier, you can upload your files and when parrot hears that question on the screen recommends the answer. Which i believe very handy.If you want to make it smarter and dont mind cloud you can hook up your Claude API and Deepgram account, that gives super powers to Parrot. You still keep it locally but eventually you need to make API calls.So local or cloud totally up to you.Then after months of thinking i have decided share it as open source and this is my first open source project, so please go easy on me :)No account, no membership, open source and free.Code: https://github.com/turantekin/ParrotThis project made me so excited and really would like to share with you guys so i hope can help you too. Again, i am new so go easy on me and if you have any feedback i would like to hear.Thank you Uygar
Hacker News1mo agoToolAI
Ledge.sh – Runnable Markdown Notes
Hi HN,Ledge is a Markdown notebook that runs shell commands, code, SQL, etc from inside your own notes.I built Ledge because I spend much of my day copy/pasting commands from my notes into the terminal. I was inspired by how much cmux helped me organize my terminals - but there was still a split brain between my notes and frequently run commands. I've been daily-driving it for the past few weeks and use it for deploys, API calls, smoke tests, etc.Ledge runs your real shell just like a terminal app and can be hosted locally or remotely over SSH using ledge-server. I've been building it since July and have recently added support for all the major platforms: Mac (Silicon), Windows (WSL required), Linux, iOS, Android. Mobile devices require SSH access to a ledge-server and Android is still in beta and looking for beta testers (see link on website)!It's built on Bun and Electrobun and is free and open-source. Feel free to review the code and contribute at https://github.com/ledgesh/ledgeFeedback is very much welcome. Any must-have features that are missing?
Hacker News1mo agoToolAI
HN.watch – Videos of all Hacker News posts
Hi HN, I’m Per, founder of Scrimba (YC S20). We’ve spent the last decade teaching people how to code with an HTML-based video format. We’ve now plugged an LLM into it, so that people can create explainer videos about anything. It’s called “Scrimba Explain”.To demo this technology for Hacker News, we built HN.watch. It’s like HN, but with explainer videos instead of articles. We create them on-the-fly the first time someone clicks on a link.While there are obvious visual drawbacks of using HTML instead of diffusion models, there are three big benefits: - Speed: Much faster to generate than pixel-based videos (just a few seconds from click to playback) - Cost: Our cost per video is ~$0.04. (Excluding image generation, which some videos utilize. Quickly blows up the cost) - Easy editing: the above benefits also make AI-assisted editing cheap & fastOur hypothesis is that if video creation goes from “dollars and minutes” to “cents and seconds”, a bunch of new use cases will be unlocked. Here are some we see already: - A video explanation of every single Pull Request (we do this internally) - Give every page in your internal/extrernal docs a video - Turn a complex article into a video in ~4 seconds (via our Chrome extension) - Course creators can quickly draft lessons before recording the real thing - People also create a lot of personal stuff stories for their kids, wedding invitations, birthdays, etcThe stack is based on an open-source programming language (Imba) created by our CTO, Sindre Aarsæther. It compiles to JavaScript, so it interoperates fully with the npm + node ecosystem. You can learn more here: https://imba.io/We’ve also built our own sync engine (OP), and a context management system for agents (Q). We feared this would make the LLMs struggle when writing code for us, as neither is in their training data (there’s very little Imba in there too). However, we’ve been pleasantly surprised to see that LLMs actually are really good at our
Hacker News1mo agoToolAI
NSL – WSL for Linux
One of the things that Windows really got right is WSL2. I drive an atomic Linux distro for daily use, but wanted a way to develop with multiple different distros with that same WSL UX. NSL is my answer. It is a faithful reproduction of the developer experience, powered by a single VM that hosts one or more systemd-nspawn containers with your development instances. Host file edits and port sharing come along for the ride, just like WSL. Take a look and tell me what you think... It's yet another step in my long journey to keep my host installation free from all the changing and breaking dev dependencies that force a reinstall every few months.
Hacker News1mo agoToolAI
Vespper (YC F24) – SOTA Docx MCP
Hey HN! We're Dudu and Topaz from Vespper (https://vespper.com). Vespper is an MCP that lets AI agents efficiently edit Word documents, powered by our fine-tuned model. It's currently 3× faster, 2× cheaper and more accurate than the closest alternative. Check out an overview of how the product works here: https://youtu.be/odKxsgPjzzwWe came to work on this problem after spending a year building an AI document editor for pharma companies. Before that, Topaz(myself) was a senior SWE at Snyk, working on distributed systems, and Dudu was a deep learning engineer at Viz.ai, building computer vision models for stroke detection. Our editor helped pharma companies generate regulatory documents (e.g. CSRs) to speed up their submissions. Initially, the output was Markdown, displayed in a WYSIWYG editor. However, users preferred working with their own Word templates. That's when the problems began.AI agents aren't great at editing Word documents. A Word document is a zip file of verbose XML files following the OOXML spec. Even "small" changes require backflips, for example: adding a numbered list requires creating an entry in numbering.xml with a fresh ID and linking it back in document.xml, bolding a sentence requires splitting it into 3+ run elements. The list goes on.This makes editing the zip directly (unzip + grep + sed) a bad idea for agents because they burn a lot of time + tokens on these mechanics. In practice, today's tooling falls into roughly three categories. You can let the agent write code against low-level libraries like python-docx or the Open XML SDK, you can give it an MCP with opinionated editing tools (SuperDoc, Office CLI, Adeu, etc), or you can round-trip the file through Markdown/HTML with something like pandoc/mammoth.js. None of them really work. The first two categories still burn the agent's context on Word mechanics instead of the task at hand (MCPs also introduce a new DSL
Hacker News1mo agoToolAI
AI tools unavailable in Haitian Creole language
People in Haiti lack accessible AI platforms for learning, creating, working, and building in their native Creole language, limiting AI adoption in non-English speaking regions.
X1mo agoToolAI
Building a Markdown editor for Mac, iOS and web
I've spent the last 10 months building Beauty: a local, powerful greco-roman Markdown editor where the page looks finished while you type. It runs on the Mac, iPhone and in the browser, and the browser version opens without an account: https://app.markdown.beautyI was frustrated with writing apps, and like most I used Apple Notes. After 2+ years immersed in Cursor, Codex and CC, I wished some interactions existed in a text editor.The Markdown renders in place as you write it, including tables (which you can drag and resize without lining up pipes), KaTeX math and Mermaid diagrams. A lot of thought went into typography and the writing experience.Your notes on the Mac are plain .md files in a folder with images next to them, so they open in anything else too. Each note keeps 200 versions. If you share a note with a link it syncs peer to peer over WebRTC while someone has it open, and the server that connects you only ever sees a random room id.Most of the ~1,000 commits went into the editor itself. It's a contenteditable, and I don't let the browser restructure the document at all, which was a lot more work than I expected. Building this to be performant and cross-platform required some novel approaches, which I'll later open source. The editor is free with no account or ads, and there's an optional subscription for sync through your own iCloud.https://www.markdown.beauty
Hacker News1mo agoToolAI
PaperMono, e-ink fridge magnet shopping list with mobile web page
Fridge magnet shopping list on an M5Stack PaperMono (ESP32-S3, e-ink touchscreen), synced with a phone web app over Wi-Fi. Works offline, ~2,400 lines of C++.This is a great wee device! BLE, Wi-Fi and LoRa on one board makes it flexible for way more than a shopping list, this is just what I built first. This firmware only uses Wi-Fi.Fully vibe-coded with Claude Code, I didn't hand-write this. I wanted to see how Claude would get on with building something useful for a new hardware device.As it's stuck to the fridge it's easy for the family to use, and it's already being used day to day which I'll admit is a first for a home built project like this
Hacker News1mo agoToolAI
Ekselio – Loveable for finance workflows (local first)
Hi everyone, I am KD - Back in my college days, I dabbled with coding, learned the basics, HTML, CSS etc. but somehow I ended up in Finance which consumed the next 20 years. Then, during covid I picked up coding again, learned react, typescript, etc - even built a rudimentary site - and then came the chatgpt moment, followed by Claude etc.So, as a side project, considering that I had spent 20 years in finance and M&A I started building Ekselio, loveable for finance workflows.Differently from other vibe coding tools, this is local first - meaning the workflows are orchestrated by the LLM based on the file schema but then the execution happens in the browser - the cool thing is that you can see the canvas, the previews of each node, and the code base (SQl) - see the Canvas tab.That said I did include a home tab, which is more of a visual, mainly because people think it's easier to relate to how folks operate, although I know finance and accounting folks usually like the workflow and full transparency - the other thing is that once you create a workflow you can save it and then next month,you can just click Run and it runs it again with no LLM tokes involved - so the infrastructure is minimal.Everything can be exported in an excel data pack with M code in case someone would want to reproduce - think Audit ready. And, I also connected to quickbooks on line so someone can directly pull in and manipulate their financials live - I also connected to Fred which carries literally 800k series in terms of macro economic data - i.e. oil prices, inflation, CPI etc and finally to stocks.I am hoping people find holes in the architecture or any ideas to make this flow a little bit better. It is in progress so if you run into bugs let me know. Considering that I am using my own API key, I have put a wall in case people overuse it but DM me if you would like to continue using, I can turn off the wall for you - hoping to keep that to a small number of users if I can and wil
Hacker News1mo agoToolAI
AI writing detection tool
Tool to detect whether text was generated by AI vs human-written. Growing need as educators, publishers, and content teams struggle to identify AI-generated content reliably. [Google Search suggestion]
Google Trends1mo agoToolAIUS
A Claude Code skill to analyze your chess games
Hello HN,It started as an experiment: can Claude play chess properly if it uses vision instead of PGN notation? Somehow it can.The next experiment was to see whether Claude + Stockfish could explain a game. Somehow it can too.A few sessions later, I had a system that takes my live audio notes (or text, for that matter) and a vague instruction like "analyze my last lichess game", and gives me a commented video of the game. The result is not perfect and it takes time to deliver (an hour or so), but for me it is a much more pleasant and memorable experience than clicking around Stockfish branches. It burns tokens, so make sure you have enough quota. From the session logs, the last analyzed game would have cost around $15 at API prices.The fact that it reflects on my own thinking during the game makes it interesting from a teaching point of view, so I thought it was worth sharing.
Hacker News1mo agoToolAI
Reladraw – A diagram language where you decide where to place things
I love making diagrams to help understand, plan, etc. However, the options are (A) auto-placement languages like Mermaid or Graphviz (which don't let me decide how the diagram looks), or (B) software like Draw.io which are powerful but are very time consuming (and inefficient for agents to manipulate).I wanted to have the benefits of both, where you can define a diagram in a diagram language, but also retain a high degree of control over what the diagram looks like.I also wanted this to work well for humans and agents.On the Github link, there's a playground where you can try it out without installation. There's also instructions for a simple npm install and for installing a skill you can use with Claude or other agents.
Hacker News1mo agoToolAI
Is it impossible to disable Siri on macOS 27?
I disabled Siri in the Settings app, and then further disabled Siri and its extensions in Settings > Screen Time.Next, I disabled all Siri-related and Apple Intelligence-related services as per https://github.com/OleksandrKrupko/mac-os-debloatBut after restarting, I could still see a "Siri" process running in Activity Monitor. Upon further inspection, I found that it is the "Siri AI.app".These "features", together with Apple's Personal Context sound very similar to Microsoft's Recall.Does anyone have the same experience? Is it really impossible to fully turn off Siri, Siri AI, Apple Intelligence and all related services and features on macOS 27?
Hacker News1mo agoToolAI
Who's still keeping a DOS machine up because the business depends on it?
Do you currently work with or know anyone who is still using:* dBase/Clipper/CLARION/Paradox/other DOS RAD environments on period hardware to run business processes?* CNC mills/spectrometers/microscopes/other industrial instruments controlled by ISA cards (either bespoke or standards like GPIB)?* Anything with a parallel port dongle?If so, I'd be very interested in hearing your experience here, or feel free to send me an email at the address in my profile. I'm not trying to sell anything, just doing some research for an idea around keeping these going on modern hardware.
Hacker News1mo agoToolAI