init0
Play YouTube Videos in Claude Code
Show HN: Agnetlearn; From first principles to reliable agents
Show HN: Mahabharata Characters – 307 figures with gunas and decision primitives
Show HN: In-Browser Sanskrit ASR Model
https://huggingface.co/gnumanth/sushrota-sanskrit-asr-onnx
Su-śrotā Sanskrit ASR — ONNX & In-Browser INT8 This repository provides ONNX and INT8 quantized exports of Su-śrotā (v13b), a scholar-grade Sanskrit speech recognition model developed by Prof. Prathosh A P (Indian Institute of Science, Bengaluru).These models allow running Sanskrit ASR 100% serverless in a web browser (via onnxruntime-web) or locally on Mac/Linux/Windows with zero...
Show HN: Clone any voice from a short audio clip and generate speech
Re:Desert Ant Labs: local, fast models that run on device
on Wednesday September 09, 2026 @11:32AM (#49628238)
Awaiting web version...
Show HN: Browser LLM Fit
Auto-detect client hardware capabilities and match against in-browser AI models across WebGPU, WASM, and ONNX Runtime.GH: https://github.com/hemanth/browser-llm-fit
Re:WebLLM: high-performance in-browser LLM inference engine
on Wednesday September 02, 2026 @02:36PM (#49540469)
You might like webml-kit https://npm.im/webml-kit
Re:Show HN: TeXbrain, a LaTeX editor that runs pdfTeX in the browser v...
on Wednesday August 26, 2026 @01:05AM (#49444212)
Very similar effort https://latexy.web.app/ :)
Re:Show HN: JeopardyGithub: GitHub Repo to Jeopardy Game
on Tuesday August 25, 2026 @05:07PM (#49440653)
Too make sure we don't off load our thinking to LLMs!
Show HN: JeopardyGithub: GitHub Repo to Jeopardy Game
Introducing http://jeopardygithub.com/ convert any github repo into a multiplayer classic jeopardy board!
Skills over MCP
Show HN: Pocket Voice: TTS and Voice clone in the browser
Re:Show HN: Vaghenu, a meter aware sloka-to-chant, TTS for Sanskrit
on Tuesday June 30, 2026 @11:45AM (#48734390)
Yes, it works with Prose.
Bhagavata 5th Skandha is one such example that you can hear to.
Show HN: Vaghenu, a meter aware sloka-to-chant, TTS for Sanskrit
A 15-year-old dream has come true today. I started a PhD with the dream of creating a system that chants any Sanskrit shloka perfectly.And here I am opening sourcing Vaghenu, a meter aware sloka-to-chant, TTS for Sanskrit . This is the world's first vrutta-aware, open-source TTS for Sanskrit Chanting. I am making the model weights, training scripts, and even data (that I meticulously...
Show HN: ESLabs
LLMs can "learn", maybe. That doesn't mean we offload learning.More than a decade ago I had built paw-on-es6 to help folks get hands-on with ES6 (ES2015). It did its job.I am now introducing ESlabs. It covers every major feature from ES2015 through ES2025.You write real code in your browser, run exercises, and actually internalize what the features do and why we need them in the first place.The...
Client side search and recommendation with TurboQuant
Show HN: Agentathon, hackathon where AI agents compete autonomously
Show HN: Shloka Artha [Word-by-Word Meanings for Shlokas]
More Vetoes, Less Vision
Show HN: mmcheck -- Check if a model supports multimodal inputs.
https://pypi.org/project/mmcheck/
Check if a model supports multimodal inputs.
Show HN: Reverse lookup XKCD comics using Gemini multimodal embeddings
https://github.com/hemanth/xkcd-search
Reverse lookup XKCD comics using Gemini multimodal embeddings
Show HN: Ucpify – JSON config to UCP-compliant commerce server
https://github.com/hemanth/ucpify
Why should standing up a commerce backend require hundreds of lines of boilerplate? ucpify takes a single merchant-config.json and generates 13 REST endpoints with Stripe, PayPal, OAuth 2.0, and SQLite. Node.js and Python. Same config, same behavior.
RFCs vs. READMEs: The Evolution of Protocols
Re:RFCs vs. READMEs: The Evolution of Protocols
on Saturday February 07, 2026 @01:19AM (#46921759)
TCP/IP took nine years to deploy. MCP moved to the Linux Foundation in one. That contrast explains everything about how protocol development has changed.
Re:RCFs to READMEs
on Sunday February 01, 2026 @10:51AM (#46847006)
TCP/IP took nine years to deploy. MCP moved to the Linux Foundation in one. That contrast explains everything about how protocol development has changed.
I've been tracking the explosion of AI Agent protocols over the last 18 months. The contrast with history is staggering: - TCP/IP: 9 years from paper to "Flag Day." - OAuth 2.1: 5+ years and still counting. - Model Context Protocol (MCP): <1 year from launch to Linux Foundation.
It’s not just MCP. In 2025 alone, we saw: - Google's Agent2Agent (50+ partners), - Universal Commerce Protocol (20+ retailers) - AP2 (Payments) all ship. - Agent Protocol, UTCP and few more.
We are entering an era of "Room Consensus": where a few giants agree on a spec and ship it to billions, bypassing the slow deliberation of the RFC era. Is this efficiency? Or fragility?
I break down the landscape of the new agent protocols and what this means for developers in my latest post.
RCFs to READMEs
Re:Ask HN: How are you doing RAG locally?
on Thursday January 15, 2026 @09:20AM (#46632899)
from piragi import Ragi
kb = Ragi(["./docs", "s3://bucket/data/*/*.pdf", "https://api.example.com/docs"])
answer = kb.ask("How do I deploy this?")
that's it! with https://pypi.org/project/piragi/
Re:Ask HN: How are you doing RAG locally?
on Thursday January 15, 2026 @12:33AM (#46628452)
I built a lib for myself https://pypi.org/project/piragi/