ngkw

↗ HN profile [ 77.5 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. TL;DR: I have a hunch that demand for classic RAG (embeddings + vector DB) will shrink. Reasons: 1. Embedding ops cost (re-indexing, freshness) is high. 2. LLMs are getting good at iterative query expansion over plain…

  2. Hi HN, I built *ReProm* ([GitHub](https://github.com/keisukenagakawa/reprom)) as a simple command-line tool for anyone who’s ever needed to share a bunch of files (and their folder structure) with AI tools like ChatGPT.…

  3. Show HN: My Hacker News (myhackernews.ai)

    Hi HN, I've built a service that uses LLM (mainly Claude 3.5 Sonnet and GPT-4) to curate and personalize Hacker News content. Key features: - AI-driven content selection from HN - Personalization based on your interests…

  4. Hello HN! I'd like to share a very small project I've built called "Copy from URL" (https://copyurl.io). It's a minimal tool designed to solve a common frustration when working with AI chatbots like ChatGPT or Claude.…

  5. I've been gathering insights from AI developers on the essential tools and frameworks they rely on for their projects. I'd love to hear from the Hacker News community too—what are your go-to AI tools and frameworks? For…

  6. Hi HN community! I am excited to share my recently upgraded website, DebateDevil (https://debate-devil.com), an AI chat platform designed to help users practice debates, challenge their ideas, and expand their thinking.…

  7. We're excited to share our latest creation, DebateDevil (https://debate-devil.com), an innovative AI chat platform that pushes the boundaries of AI-facilitated conversations. Test your ideas, refine your arguments, and…

  8. Show HN: ChatGPT (Funny) Hack (chat-gpt-hack.vercel.app)