streaming-responses

A universal, client-side AI prompt engineering tool that enhances your prompts using local or cloud-based AI models. Transform basic prompts into detailed, professional-grade instructions without sending your data to third-party servers.

A production-ready, self-hosted RAG (Retrieval Augmented Generation) chatbot widget built with Next.js, Redis vector storage, and multiple AI model fallbacks. Perfect for embedding into portfolio websites or any web application.

Personal site built with Next.js 16, React 19, TypeScript and Tailwind, including a streaming chat assistant on NVIDIA NIM llama-3.1-8b with per-IP rate limiting and prompt-injection guards

Full-featured AI Chat Interface featuring real-time response streaming, customizable AI system prompts, rich markdown & code formatting, dark mode, and conversation history.

A simple Express.js AI chatbot app with streaming responses, multi-provider support, model switching, session-based chat memory, rate limiting, and optional basic authentication.

Production-oriented RAG backend in Go with hybrid retrieval, model fallback, and ingestion pipelines