experiment - Moozonian Search

github.com• Dec 7, 2025• 1 min read

Show HN: I built an open-source Python alternative to LabVIEW for my Physics PhDHi HN, OP here.I’m a physics researcher working on cryogenic transport measurements. I got frustrated with the rigidity of proprietary drivers and the cost of LabVIEW, so I spent the last 3 years building a Python suite to replace them.It wraps SCPI commands for common instruments (Keithley 2400/6221, Lakeshore 350) and uses a multiprocessing architecture to ensure the GUI never freezes during long data acquisition loops—a common issue I faced with single-threaded scripts.It’s currently powering real experiments at UGC-DAE CSR in Mumbai. I’m looking for feedbacks.Repo: https://github.com/prathameshnium/PICA-Python-Instrument-Con...

github.com• Dec 4, 2025• 1 min read

Show HN: RIMC – An Alpha-Drift Framework for Finite-Speed Learning MarketsI published a small theoretical project on GitHub and would appreciate feedback from anyone working with real quant models.RIMC (Recursive Intelligence Market Cycle Hypothesis) is not a trading system or a full asset-pricing model. It’s a hypothesis for treating alpha as structural drift that arises from finite-speed learning and observation delay—rather than as a leftover regression residual. The goal is to write that structure explicitly in equations.The repo is mostly text with a small sample simulation. Think of it as a dynamical-systems-meets-finance thought experiment.I’d be interested in reactions such as:• Does this framing of persistent alpha make sense in practical quant terms? • Are there existing models/papers that do something similar? • How does this compare with your experience handling delay or learning effects?Any feedback—critical or supportive—is welcome.GitHub: https://github.com/rimc-lab/RIMC

m-verse.com• Dec 4, 2025• 1 min read

Show HN: Letters From the Earth – daily climate data turned into narrativeHi HN,I’m sharing a small experiment called Letters From The Earth. The idea is simple: instead of presenting climate and environmental data as charts or stats, the system converts daily signals into expressive, first-person prose — “letters” from regions on Earth describing how they feel.What it does- There is an interactive globe covering 103 locations worldwide.- Every day the engine collects signals like temperature, humidity/moisture, wind, air clarity, and anomaly deviations.- These raw values are normalized, compressed into a feature vector, then passed through a biome-specific, emotionally aware translation layer. Based on that it produces a “letter” written in the voice of the biome (forest, tundra, desert, mangrove, etc). No jargon, no statistics — just sensory, atmospheric narrative.- On the site you pick a location, read that day’s letter, and if you want compare with the raw data. There are filters by biome type, severity (anomaly), region and date so you can browse previo

docs.trysoma.ai• Dec 2, 2025• 1 min read

Show HN: I built an open-source Rust/TS AI agent runtime with a Next.js-style DXLink: https://docs.trysoma.ai/For the past ~9 months I’ve been building Soma, an open-source AI agent & workflow runtime written in Rust, with a TypeScript SDK (Python coming soon). It’s not a framework; it’s meant to sit underneath whatever agent/tooling code you already write (Vercel AI SDK, LangChain, custom code, etc.). It provides features around your framework + a better DX for building agents. I’ve tried to take a Next.JS model: open-source, good DX, self-deployable.I originally set out to build a vertical back-office/operations product for SMEs. I needed a way to coordinate multiple AI agents, workflows, and SaaS integrations (Xero, Gmail, Slack, etc.) behind a single chat interface — something that behaves like “one employee”, even though it’s made of many agents under the hood.It’s very early. The core runtime, MCP server, and TypeScript SDK are usable, but the multi-agent routing layer is still experimental. I’d really appreciate technical feedback, critiques, or use cases t

github.com• Dec 1, 2025• 1 min read

Show HN: PICA – open-source Python Alternative to LabVIEWHi HN,I built PICA (Python-based Instrument Control and Automation), an open-source software suite designed to handle complex lab measurement instruments and automate experimental workflows. It is a lab-tested, physically accurate alternative to proprietary solutions like LabVIEW.What it does: PICA provides a professional dashboard to orchestrate high-precision hardware (Keithley SourceMeters, Lakeshore Controllers, LCR Meters) for materials science research. It automates protocols like I-V sweeps, Resistivity vs. Temperature, and Pyroelectric measurement loops.Key Features: Open Source & GUI: Includes a full graphical interface for setting parameters and running experiments, making it accessible to researchers without coding experience. Process Isolation: Uses a multiprocessing architecture to run instrument control loops in isolated processes. This ensures that hardware timeouts or driver crashes do not freeze the main UI. Live Visualization: Features real-time plotting utilities to

news.ycombinator.com• Nov 25, 2025• 1 min read

I built Jelly: an SSH-only social space (no passwords, no emails)Hey everyone. I’ve been building a small experiment and wanted to share it.Jelly is a social space you join entirely through SSH. No web, no app, no login flow. Your SSH public key is your identity.I wanted something that felt like early internet communities. Simple, no algorithms, no rage-bait, no scraping, no AI content. Just people talking in real time from the terminal.What it currently doesJoin with: ssh jellyssh.xyzIf you don't have a key: ssh-keygen -t ed25519Public keys auto-create accountsGlobal feed with real-time updatesHashtags become channels (top 5 active topics)ASCII profile pictures + profile themesGuestbooks / Top 8 (MySpace nostalgia)TUI built with Bubble TeaSSH server powered by Wishsqlite backend, no tracking or analyticsNo passwords, emails, or personal data storedIt’s very early and rough around the edges, but stable enough to try. Still tightening up input handling, rate limiting, and crash-proofing.If you check it out, I'd love your thoughts/feedback. It’s just

resonantlabsai.github.io• Nov 20, 2025• 1 min read

Show HN: TDS Compass – AI prompt for your communication styleHi HN,I’ve been experimenting with a small tool called TDS Compass and would love feedback.What it isA 1-minute, 8-question quiz that maps your communication style onto two axes:Structure (S) – how much you prefer scaffolding, steps, and clear organizationRelational (R) – how much you care about tone, warmth, stories, social contextThose scores map into 16 “zones” (A1–D4).For each zone, the app generates:A short explanation of how you tend to communicateA system-style prompt you can paste into ChatGPT / Claude / etc. so the model adapts to that styleLinksQuiz: https://resonantlabsai.github.io/tds.compass/quiz.htmlHome / overview: https://resonantlabsai.github.io/tds.compass/Source: https://github.com/resonantlabsai/tds.compassAll static, no backend, no login. Just HTML/CSS/JS + JSON.Why I built itA lot of “prompt guides” treat AI purely as a tool to optimize. I wanted something that:Helps people understand their own information + relational preferencesTunes AI interactions so they feel

ztgiai.pages.dev• Nov 16, 2025• 1 min read

Show HN: ZTGI-AC – An AI that checks its internal stability before answeringHi everyone,This is a small experimental AI project I’ve been building called *ZTGI-AC*.Most LLMs generate an answer immediately, but ZTGI-AC does something different: before responding, it runs an internal stability check.It evaluates: • risk • jitter • dissonance • SAFE/WARN/BREAK modes • INT/EXT gating (self-monitoring loop)Only after the internal signals stabilize does it generate a reply.This project explores whether self-evaluation loops can reduce chaotic or unstable outputs in LLM-like systems.*Demo:* https://ztgiai.pages.dev (Non-commercial, early prototype.)I’d love feedback from the HN community — especially around: • whether self-monitoring loops are meaningful, • potential improvements to stability metrics, • and how this idea compares to classical alignment approaches.Thanks for taking a look!

zenpaint.org• Nov 15, 2025• 1 min read

Show HN: ZenPaint, a pixel-perfect MacPaint recreation for the browserI've been recreating the original MacPaint in the browser on and off for a few years. It's still alpha quality, but I'm finally ready to share it more widely.The goal was pixel-perfect accuracy, so I spent a lot of time with Atkinson's original QuickDraw source code, emulators, and my iBook G3 to get details like font rendering and the shape tools exactly right.Some technical notes:- Font rendering was surprisingly tricky; understanding the original pipeline's quirks took lots of experimentation, and avoiding canvas smoothing/aliasing required careful handling.- Written declaratively with React; performance is kept reasonable with a buffer pool and copy-on-write semantics. - You can share links to artwork from within the UI.E.g.: https://zenpaint.org/#p=KQumBQ5x- Mobile support was not considered here (for obvious reasons). It might still be usable on a larger phone or tablet but I have not tested this at all.There's something magical about making art within MacPaint's constraints: the

huggingface.co• Nov 14, 2025• 1 min read

Show HN: Cleaned 2.7M French Wikipedia JSON articles (full dataset)"I’ve been working on this project mostly out of curiosity and because I like cleaning messy data. I downloaded the full French Wikipedia dump (the raw XML + wikitext one) and built a script that extracts every article and turns it into a clean JSON file.Nothing fancy — just a pipeline I wrote myself that:reads the XML streampulls out each pageremoves wikitext and leftover markuprebuilds the sectionsparses the infobox into a real JSON objectextracts categories, links, etc.and then saves everything as one JSON file per articleThe result is around 2.7 million JSON files, each representing a single Wikipedia article in a format that’s directly usable for NLP or LLM experiments.This wasn’t meant to compete with existing datasets — I just wanted to understand how to process the dump properly and build something clean from scratch. Since it turned out well, I’m sharing it in case it helps anyone.I’m also running the same process on the full English dump (around 6.2M pages). Still in progress.

kinkora.fun• Nov 10, 2025• 1 min read

Show HN: Exploring the boundaries of AI-generated creativityHi HNI’ve been experimenting with AI image and video models for a while, and one thing I kept running into was creative limits — many tools restrict what can be generated or explored.So I built a new platform focused on unrestricted creative exploration. It allows users to experiment with AI-generated images, videos, and audio, with full freedom to test different models, styles, and concepts.The goal isn’t to shock or provoke — it’s to give creators space to explore ideas that are often filtered out elsewhere, and to study how AI can represent human imagination in its raw, unfiltered form.We’re starting with a few models and will expand to support more in the future. Still early days, but if you’re interested in the boundaries of AI media generation, I’d love your thoughts and feedback.

monkeyai.cc• Nov 10, 2025• 1 min read

Show HN: MonkeyAI – Curated creative templates for AI image generationHi HNI built MonkeyAI — a site that curates creative templates and examples for AI image generation, designed to help artists and designers find inspiration faster.When experimenting with different image models, I noticed that generating great visuals often depends on how the prompt is structured — the tone, subject, and composition details matter a lot. But most prompt libraries are either overwhelming or lack creative direction.MonkeyAI focuses on quality and clarity. Each template comes with:A visual preview of the resultThe exact prompt structureContext tags for mood, style, or themeA clean interface to explore ideas visuallyThe goal is to make AI creation more intuitive and inspiring, especially for people who think visually — designers, illustrators, and creative makers. I’d love your feedback on:What would make this more useful for your creative process?Would you like to see model-specific templates or open submission from users?

github.com• Nov 5, 2025• 1 min read

Show HN: I was tired of ROS2, so I rewrote it in RustEvery once in a while, I hear developers complain about how working with ROS2 frustrates them, or students who find the barriers to understanding ROS2 too high. Undoubtedly, ROS2 was originally designed in the lab for the prototyping phase.However, being widely adopted has turned its design into a messy blob that is not only "thick" - and not in a good way - but also too tangled. Therefore, I came up with the design for HORUS -Hybrid Optimized Robotics Unified System.At first, I was just experimenting to see whether building a system that respects both the versions and dependencies of low-level packages like drivers, and high-level packages like ML/robotics algorithms would be that hard - unlike in ROS2. Surprisingly, with current modern technology, we can do a lot to make this vision come true. ROS2 users are always looking to avoid "reinventing the wheel" by trying to pick whatever exists in that tangled web and sticking it into their project. This workflow was good, at least in the

github.com• Nov 4, 2025• 1 min read

Show HN: Nallely a modular reactive Python framework for custom MIDI instrumentsHi HN! I'm Vince. I built Nallely, a modular reactive Python framework for creating custom MIDI instruments by patching signal-processing modules together, like a modular synthesizer for controls systems. Nallely focuses on real-time, thread-isolated, reactive behavior, letting you experiment with emergent behaviors.Demo video: https://www.youtube.com/watch?v=rbMnKAdqAVI building a patch from scratch and hot-debugging a running instance near the end.Key features:* Visual patching interface for connecting reactive modules (neurons),* Extensible via Python API, WebSocket, and/or code generation,* Integrates any input source (MIDI, webcam, ...) to control synthesizers.# Yes, but why?Existing software/libraries that proposes MIDI manipulation are powerful but not friendly to live experimentation. They are low-level, hard to rewire on the fly, and often heavy for embedded or headless setups. I wanted a system that could also evolve dynamically where modules could be patched, hot-swapped, an

news.ycombinator.com• Nov 3, 2025• 1 min read

Ask HN: Is someone building an AI design agent?I’ve been thinking about the future of design workflows and wondering if anyone here is working on an AI design agent, not just another image generator, but something that actually understands design intent and executes creative direction end-to-end.Something that could: - Take prompts like “create a campaign for a new streetwear drop” and generate cohesive brand visuals (logo, apparel mockups, ad creatives, etc.) - Learn a brand’s visual identity over time - Suggest new content ideas or remix existing assets - Work more like a creative assistant than a “prompt in, image out” toolBasically, a creative automation platform that behaves like a designer who knows your brand, not just a diffusion model with style filters.Is anyone building something like this, or thinking in that direction? Would love to see examples, prototypes, or even open-source experiments.Also curious: what would you want an AI design agent to handle for you? Mockups? Social posts? Full brand systems?

news.ycombinator.com• Oct 27, 2025• 1 min read

I thought cheap products get more users. I was wrongThis month, my SaaS QRAnalytica.com hit a new milestone: $1,046 in revenue.For a solopreneur building this after my day job, this number means the world. But the path to get here was paved with a massive mistake I kept making.When I launched back in December 2024, my logic was simple: make it as cheap as possible. I priced it at $99. I thought, "The lower the price, the more people will sign up, right?"Wrong. So wrong.The first dollar didn't even come until February 14th, 2025. For months, I was dealing with low-value sign-ups and felt like I was spinning my wheels.The turning point was a random LinkedIn post by Marc Lou. He talked about pricing based on value, not on being the cheapest option. A lightbulb went off.I decided to run an experiment. I started raising my prices.From $99 to $149... a few more signups.From $149 to $249... things started to change. I attracted a different type of customer—one who understood the value and was ready to invest in a real solution.I even tested $3

launchy.iridesk.com• Oct 26, 2025• 1 min read

Show HN: I made a brutalist closer-to-retro directory for new product launchesi built Launchy with the help of an open source base as a small experiment in bringing back the feel and vibe of old-school web directories, but with a focus on new product launches.the goal was to see if directories could feel more human and less like a modern simplistic vibe everyone seems to be going for these days.you can browse launches, submit your own, or just enjoy the aesthetic.i'm curious what you think, does the minimalist and brutalist throwback design still appeal to today's builders, or are the modern clean and simplistic approaches better?(p.s. there's also an unscrambling game hidden in the footer)

news.ycombinator.com• Oct 24, 2025• 1 min read

Show HN: The Σ-Manifold ManifestoThis project explores the connection between *the linear structure of text* and its *emotional-aesthetic impact*. We identify *five fundamental relations* between consecutive sentences — labeled *A–E*. Each represents a shift of *subject-object*, i.e., a transformation of perspective and agency.When texts grow longer, these relations form *sequences* — and from the infinite combinatorial space, *eight stable patterns (Σ₁–Σ₈)* emerge empirically. Each pattern correlates with a distinct *semantic and emotional field* — cathartic, heroic, meditative, humorous, and so on.This allows us to instruct an LLM not through semantic prompts (“write a story about…”), but through *structural commands* — e.g., generate a narrative following sequence Σ₅ (Tragic Counterpoint). You can experiment with these archetypes directly here: [Narrative Generator](https://a2tg9zwayjuqzcpdznklve.streamlit.app/~/+/#narrative-...) or [via python](https://github.com/mihendr/Echoes-of-autonomy/blob/main/TEMA...)Intere

ai-video-gen.org• Oct 24, 2025• 1 min read

Show HN: AI Video Generatorai-video-gen.org is a newly launched AI video generator currently in its promotional and trial phase. This early rollout invites creators, marketers, and teams to test the platform at lower prices while collecting feedback to sharpen features and reliability. The aim is to validate workflows, output quality, and pricing before a full-scale launch, with an emphasis on fast iteration and user input.What to expect during this phaseAffordable entry: Generating videos now comes at reduced prices to encourage experimentation and rapid content production across campaigns and experiments.Risk-free trial: The platform prioritizes a low-barrier experience, making it easy to try different concepts without heavy upfront commitments.Free first-generation credit: New users receive a one-time free video generation, enabling immediate hands-on evaluation of capabilities and output quality.Billing and policies you should knowLow-cost testing: The current stage prioritizes accessibility; pricing is desi

github.com• Oct 22, 2025• 1 min read

Show HN: Confidential AI browser extension – alternative to OpenAI AtlasHi HN,We published our experimental browser extension for Chrome just a few hours before OpenAI announced its Atlas browser. Ours takes a different approach: privacy first.Atlas makes “memory” and cross-tab context a core feature. OpenAI acknowledges the privacy concerns and promises users full control over what gets stored and sent to the cloud. In practice, most users won’t maintain that control.Our vision is full privacy by design, so it doesn’t matter what data flows to the cloud. The extension connects to (1) a cloud-hosted inference service (Privatemode AI) running gpt-oss-120b in carefully architected confidential-computing environments and (2) a local document store with a vector DB.The inference service leverages AMD SEV-SNP and the confidential-computing features of the Nvidia H100. The environment is designed so that we (Edgeless Systems) and no one else can access user. All data stays encrypted even during processing in main memory. The client automatically verifies the int