jina-ai

An agentic extractor for messaging platforms. Noria receives messages & URLs via chat; scrapes web data to evaluate the content against custom scoring matrices (e.g., jobs, scholarships); and sends a structured summary with a calculated match score directly to the assigned messenger number.

Ask a web page a question and highlight the answer in place, with one listwise jina-reranker-v3.5 call. No index, no vector store. Runs locally on Apple Silicon via MLX.

A modern API-driven news aggregation platform. Features real-time news, AI-powered article extraction, market watch, dark/light theme, and customizable reading experience.

Enterprise Agentic RAG Platform for Indian Law (BNS, Constitution). Features Confidence-gated HITL Web Search, Semantic Routing, and real-time Prompt Injection Guardrails.

AI-powered Discord server management bot. Understands natural language commands to manage roles, channels, members, and moderation — with multi-provider AI fallback, background job system, and two-phase execution architecture."

Production-grade RAG Document Intelligence Platform — LangGraph Adaptive RAG + CRAG, LangSmith observability, RAGAS evaluation,Qdrant hybrid search (BGE dense + BM42 sparse + RRF), jinaai/jina-reranker-v1-tiny-en  re-ranking, FastAPI + Docker

A Agentic RAG system built with LlamaIndex and Qdrant for NCCU teachers and students.

Автономный движок контент-операций на базе ИИ. Автоматизация публикации на различных платформах, создание SEO-оптимизированных черновиков и дистрибуция в социальных сетях с использованием n8n и Gemini.

Autonomous AI-driven content operations engine. Automate multi-platform publishing, SEO-optimized drafting, and social media distribution using n8n and Gemini.

Text embeddings and reranking for Swift — one protocol, providers for Voyage, Jina, Cohere, OpenAI and on-device NaturalLanguage. Back-deploys to iOS 15.

A configurable Deep Research agent and TypeScript framework for DeepSeek Harness, with web search, optional page and PDF reading, and cited reports.

Enterprise Research Skill｜企业深度研究 Skill：Evidence-grounded research for Codex, Claude Code, WorkBuddy, OpenClaw and other agents. Read full sources, cross-check claims, and produce traceable reports.

Personal developer portfolio highlighting projects, skills, and contact details with a clean responsive interface.

A text-mode web browser for the terminal: any URL → clean Markdown, with reader & AI-summary modes and Obsidian-compatible saving.

Production-grade Financial RAG for SEC 10-K filings with structure-aware parsing, semantic chunking, multi-stage retrieval, Jina reranking, and source-grounded answers

Chrome extension (MV3) that extracts links from the current page, fetches each as clean markdown via r.jina.ai, and bundles them into a single zip download

Intent-driven workflow automation bot utilizing n8n, Gemini AI, and Jina AI for protected data extraction.

Serverless client-side intelligence dashboard — real-time Hacker News radar, GitHub trends, and on-demand URL scraping via Jina AI. Zero API keys, zero backend, hosted on Vercel.

Markdown-first web retrieval skill for AI agents (Cloudflare + Jina + Firecrawl)

Image Captioning Refinement Using Deep Learning and computer-vision — Flickr8k dataset with BLIP, Jina, Qwen and Majority Voting