🛠️ All DevTools

[Other] tldraw SDK 4.0

[Other] MapSCII – World Map in Terminal

[Other] Show HN: I Parallelized RNN Training from O(T) to O(log T) Using CUDA

[DevOps] OneDev – Self-hosted Git server with CI/CD, Kanban, and packages

[Other] Show HN: KSON, a love-letter to the humans maintaining computer configuration Hi friends, I&#x27;m really excited to introduce KSON, which just entered public beta!<p>Anywhere a human is reading or editing YAML&#x2F;JSON&#x2F;TOML, KSON may be used as a more effective interface on that data. If you are such a human, we invite you to participate in this beta.<p><i>tl;dr</i> Check out the website [1], play with the online playground [2], install the library for your programming language [3], edit in your favorite editor [4], discuss and give feedback [5], contribute to the project [6].<p>(A personal note about this project: I love software. Machines made of words! Such a wonder. KSON itself, as a collection of words that both make a machine <i>and</i> explain that machine, is an expression of a lot ideas I feel really passionately about around software and our relationship to it. I&#x27;ve put a lot of love into trying to make that expression eloquent and reliable. I hope some of that comes through clearly, and I look forward to discussing this more over time with anyone who&#x27;s interested)<p>One of the key things KSON wants to say is: let&#x27;s keep everything that&#x27;s great about YAML and JSON as &quot;Configuration User Interfaces&quot;, and let&#x27;s make those interfaces more toolable, robust, and fun. Here&#x27;s some of the ways we do that:<p>- KSON is a verified superset of JSON, has native JSON Schema support, transpiles cleanly to YAML (with comments preserved!), and is likely available wherever you want it—current supported platforms: JS&#x2F;TS, Python, Rust, JVM, and Kotlin Multiplatform.<p>- KSON is also widely available in developer tools, with support for VS Code, Jetbrains IDEs, and anywhere you can plug in an LSP.<p>- KSON is fully open source, licensed under Apache-2.0, and you are invited to meet its words and tinker with how they make its machine. A lot of care, craft, attention and joy went into making the KSON project understandable and approachable for developers. We hope to see you around.<p>PS. This is an HN-friendly version of the official announcement at &lt;<a href="https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;blog&#x2F;2025&#x2F;09&#x2F;17&#x2F;introducing-kson&#x2F;" rel="nofollow">https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;blog&#x2F;2025&#x2F;09&#x2F;17&#x2F;introducing-kson&#x2F;</a>&gt;.<p>[1]: <a href="https:&#x2F;&#x2F;kson.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;kson.org&#x2F;</a><p>[2]: <a href="https:&#x2F;&#x2F;kson.org&#x2F;playground&#x2F;" rel="nofollow">https:&#x2F;&#x2F;kson.org&#x2F;playground&#x2F;</a><p>[3]: <a href="https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;install&#x2F;#languages" rel="nofollow">https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;install&#x2F;#languages</a><p>[4]: <a href="https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;install&#x2F;#editor-support" rel="nofollow">https:&#x2F;&#x2F;kson.org&#x2F;docs&#x2F;install&#x2F;#editor-support</a><p>[5]: <a href="https:&#x2F;&#x2F;kson-org.zulipchat.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;kson-org.zulipchat.com&#x2F;</a><p>[6]: <a href="https:&#x2F;&#x2F;github.com&#x2F;kson-org&#x2F;kson" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;kson-org&#x2F;kson</a>

[Testing] Fast, easy and reliable testing for anything that runs in a browser.

[Other] A complete computer science study plan to become a software engineer.

[Other] 🚀Introducing NoLimitsGPT— The AI with no limits🚀 Meet the AI that has no limits: 💬 Answering Questions 🧠 Problem Solving 👨‍💻 Content Creation 💡 Concept Explanation 🎬 Content Generation 🖼️ Image Generation 🖥️ Web Development 📰 Real-Time Information Access ⚡Workflows 🤖AI-Agents and more

[Other] Font Identifier for Mac, iOS, iPadOS & Menu Bar Identify fonts instantly with Fonti 2.0 — the font identifier & finder app for Mac, iOS & iPadOS. Detect fonts from images, organize results, explore alternatives, and use the new Menu Bar shortcut. Perfect for designers, developers, and creatives.

[Other] AI Journey Starts Here Rent high-performance GPU servers from Cyfuture to accelerate AI, ML, big data, and 3D rendering workloads. Enjoy on-demand scalability, cost efficiency, and enterprise-grade infrastructure with 24/7 support for seamless cloud computing.

[Other] Get trusted answers, instantly stackoverflow.ai is an AI-powered search and discovery tool designed to modernize the Stack Overflow experience by helping developers get answers instantly, learn along the way and provide a path into the community.

[Other] AI PM Agent for instant PRDs & user stories after meetings Stop slow sprints and high code churn. Vector transforms natural-language ideas from meeting transcripts into structured PRDs, policy-aware acceptance criteria, and developer-ready work items integrating seamlessly with Jira, GitHub, and PR review bots.

[Other] The fastest way to build mobile apps — powered by Flutter Prompt with AI, refine UI visually or dip into code if you want, all synced so you can switch seamlessly between them. Preview instantly, publish real apps and tap into Flutter’s speed, flexibility and sleek UI with production-ready code that’s always yours.

[IDE/Editor] All-in-one browser extension for web developers/designers Supercharge your web development workflow with an all-in-one browser extension that helps you inspect, edit, test, debug, and optimize websites faster.

[Other] Free online pastebin alternative Free online pastebin for sharing code snippets instantly. Secure code sharing platform with syntax highlighting for 50+ programming languages. Best pastebin alternative for developers.

[Other] Sniper Bot Accuracy for Instant Crypto Trades Our crypto sniper bot development provides an automated trading tool that enables traders to buy newly launched tokens the instant they become available, giving you faster entries, higher profit potential, and a competitive edge in the crypto market

[Other] Show HN: Made NZ's member of parliament financial disclosure data searchable New Zealand Members of Parliament are required to annually disclose their financial interests (shareholdings, directorships, consultancies, etc.) but this gets published annually as an unwieldy 80+ page PDF that&#x27;s hard to search effectively.<p>I processed structured data out of the PDF and built a searchable interface: <a href="https:&#x2F;&#x2F;open-register-of-pecuniary-interests.joshmcarthur.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;open-register-of-pecuniary-interests.joshmcarthur.co...</a>.<p>You can search across all MPs&#x27; disclosed interests by name, company, or interest type. For example, you can quickly find which MPs have interests in specific sectors or companies, filter by category or political party.<p>The data extraction was interesting - I found that a two-pass approach worked well with Gemini 2.5 Flash - one to pull out MP names and referenced page numbers, then I extracted the specific pages each MP appeared on and extracted structured data just from these pages.<p>The approach could work for similar transparency registers in other countries - most seem to publish open data as PDF, which technically ticks the box, but isn&#x27;t the most accessible format to work with. Even within NZ, I&#x27;m planning to expand the data I process to previous years, as well as processing data for local and regional councils (who have the same legal requirement to publish financial interests of council members).<p>Open sourced at <a href="https:&#x2F;&#x2F;github.com&#x2F;joshmcarthur&#x2F;open-register-of-pecuniary-interests" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;joshmcarthur&#x2F;open-register-of-pecuniary-i...</a>.<p>Tech stack: Ruby on Rails, SQLite (FTS5), Tailwind&#x2F;DaisyUI - keeping it lightweight since this is just a side project to make public data more accessible.

[Other] Show HN: Pgmcp, an MCP server to query any Postgres database in natural language

[Other] Optimizing ClickHouse for Intel's 280 core processors

[Other] Launch HN: RunRL (YC X25) – Reinforcement learning as a service Hey HN, we’re Andrew and Derik at RunRL (<a href="https:&#x2F;&#x2F;runrl.com&#x2F;">https:&#x2F;&#x2F;runrl.com&#x2F;</a>). We&#x27;ve built a platform to improve models and agents with reinforcement learning. If you can define a metric, we&#x27;ll make your model or agent better, without you having to think about managing GPU clusters.<p>Here&#x27;s a demo video: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;EtiBjs4jfCg" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;EtiBjs4jfCg</a><p>I (Andrew) was doing a PhD in reinforcement learning on language models, and everyone kept...not using RL because it was too hard to get running. At some point I realized that someone&#x27;s got to sit down and actually write a good platform for running RL experiments.<p>Once this happened, people started using it for antiviral design, formal verification, browser agents, and a bunch of other cool applications, so we decided to make a startup out of it.<p>How it works:<p>- Choose an open-weight base model (weights are necessary for RL updates; Qwen3-4B-Instruct-2507 is a good starting point)<p>- Upload a set of initial prompts (&quot;Generate an antiviral targeting Sars-CoV-2 protease&quot;, &quot;Prove this theorem&quot;, &quot;What&#x27;s the average summer high in Windhoek?&quot;)<p>- Define a reward function, using Python, an LLM-as-a-judge, or both<p>- For complex settings, you can define an entire multi-turn environment<p>- Watch the reward go up!<p>For most well-defined problems, a small open model + RunRL outperforms frontier models. (For instance, we&#x27;ve seen Qwen-3B do better than Claude 4.1 Opus on antiviral design.) This is because LLM intelligence is notoriously &quot;spiky&quot;; often models are decent-but-not-great at common-sense knowledge, are randomly good at a few domains, but make mistakes on lots of other tasks. RunRL creates spikes precisely on the tasks where you need them.<p>Pricing: $80&#x2F;node-hour. Most models up to 14B parameters fit on one node (0.6-1.2 TB of VRAM). We do full fine-tuning, at the cost of parameter-efficiency (with RL, people seem to care a lot about the last few percent gains in e.g. agent reliability).<p>Next up: continuous learning; tool use. Tool use is currently in private beta, which you can join here: <a href="https:&#x2F;&#x2F;forms.gle&#x2F;D2mSmeQDVCDraPQg8" rel="nofollow">https:&#x2F;&#x2F;forms.gle&#x2F;D2mSmeQDVCDraPQg8</a><p>We&#x27;d love to hear any thoughts, questions, or positive or negative reinforcement!