irqlevel
Show HN: Nos – a hobby x86-64 C++ OS kernel running on real KVM clouds
https://github.com/irqlevel/nos
Hi HN,I want to share nos, a hobby x86-64 OS kernel I’ve been working on.I started this mainly because I wanted to learn kernel development properly, and studying the Linux kernel feels pretty overwhelming these days. It’s huge, highly optimized, and hard to follow end-to-end. So I decided to build something smaller but still realistic.nos is written in C++20 + NASM and currently has:SMP +...
Transcribe audio files in MP3, WAV, M4A, FLAC, and OGG formats via drag and drop.
Full transcription history with searchable text.
Recording overlay, start-minimized option, and hold-to-talk mode.
WhisperClip: 300ms Voice-to-Text on macOS (Fully Local)
Show HN: Real-Time AI English Speaking Tutor
Show HN: Real-time, open-source voice assistant in Rust
https://github.com/cydanix/voice-agent
Work in progress. User pause detection may not working properly and llm/tts interruptions is not implemented yet. It's already can run and answer questions, but you need headphones to avoid echo problem (use laptop mic as input and headphones for output). Sorry, I'm not Rust pro, so the code maybe ugly, most of the code vibe-coded.
Show HN: Rust Client Library for Gradium.ai TTS/STT API
Show HN: Golang Client Library for Gradium.ai TTS/STT API
Show HN: Privacy-First AI Document Chat for macOS (Offline RAG)
https://github.com/cydanix/deepfind
I built a privacy-first Retrieval-Augmented Generation (RAG) app for macOS. It lets you search, index, and chat with your local documents directly on your Mac. Everything runs fully offline — no cloud, no tracking.The app is currently in beta and supports PDF files only. The only time an internet connection is needed is during the initial LLM model download at setup.Any feedback is very much...
Update: Replaced the site video with a demo showing the actual app behavior.
Show HN: Privacy-First Voice-to-Text for macOS
Show HN: Stream Voice-to-Text in Real Time on macOS [video]
Show HN: Fully On-Premises Inference Server for Open-Source LLMs
Show HN: AI-Powered Stock Tracking and Analysis
I’ve built StockAdvisorAI, a platform that uses AI to analyze market trends, provide trading signals, alerts, and detailed stock reports, and help track your favorite stocks. It’s in beta, and I’m actively working on adding more features, data sources, and improvements. I’d love your feedback!
Show HN: Grammar checker for text on website pages
Show HN: Codeinspect – Check source code for issues
Limitations: Only open repositories on https://github.com are available for check. Repository size must be less than 2000KB. A maximum of 10 files are selected for checking, with each file not exceeding 10KB. Reports are not saved to disk. The engine may generate false-positive issues. The service may stop working after some time.Internals: The backend uses the ChatGPT-4o API.Example...