pgvector-retrieval

AI Agent buat mantau keruwetan Indonesia. Scrape berita, analisis pakai LLM. Bertanya ke chatbot RAG interaktif.

An AI-powered compliance and procurement intelligence platform that analyzes tender documents and user queries to recommend applicable Bureau of Indian Standards (BIS) regulations using hybrid search and LLMs.

End-to-end credit decision system that combines policy/regulatory retrieval, modular agent pipelines, and LLM-assisted reasoning to produce explainable loan decisions with citations, metrics, and a Streamlit review UI. Containerized with Docker and Docker Compose for easy deployment.

Deterministic Hybrid RAG (PostgreSQL/pgvector + 3-Way RRF), Static Python AST Inspector & Closed-Loop Multilingual Diagnostic Solver

图文内容生产 —— 面向图文内容生产全流程的 AI Native 系统；图文内容生产 / AI-Native content studio — FastAPI + SQLAlchemy + Alembic, PostgreSQL/pgvector, and a static demo for image-text content workflows.

RAG pipeline for automotive product pages: Playwright ingestion, pgvector hybrid search, Ollama, FastAPI on Tata automobiles — cars, bikes, scooters and spare-parts

AI-powered career intelligence platform built with FastAPI, PostgreSQL, pgvector, and Ollama. Analyze resumes, identify skill gaps, generate learning roadmaps, prepare for interviews, and chat with an intelligent career assistant using Retrieval-Augmented Generation (RAG).

It is a concept-aware Retrieval-Augmented Generation (RAG) system for examining questions about consciousness, selfhood, identity, perception, experienced reality, and related themes across three distinct knowledge traditions

Source-grounded RAG over a medical-tourism corpus — FastAPI + pgvector (HNSW + MMR), cite-or-refuse grounding delegated to an external chat surface, React admin dashboard, LangChain ingestion.

LyricLens is a RAG-based semantic lyric retrieval system for emotion-aware music recommendation. 🐳Local 🚀Active Development

This project is a Spring Boot chatbot that lets users upload their own files and ask questions about them, using Retrieval-Augmented Generation (RAG) with a local Ollama LLM.

AMKP (Agentic Multimodal Knowledge Plane): API/SDK/MCP-first Evidence + Policy + Tenancy plane for multi-product RAG — hard isolation, Parse Ladder / TableEvidence, guarded agentic retrieve, traces & eval.

RAG - Retrieval-augmented chat API built with Python FastAPI, LangChain, an OpenAI-compatible local LLM endpoint, and PostgreSQL/pgvector

An open-source, self-hosted assistant that delegates real work to a swarm of specialist AI agents.

Tool-using agent that answers questions about SEC 10-K and 10-Q filings with verified citations. Hand-rolled loop over pgvector search, XBRL financial facts and a safe calculator. FastAPI, streaming UI, AWS.

Production-grade SEC 10-K financial RAG assistant across 25 filings (AAPL, AMZN, GOOGL, MSFT, NVDA). Features hybrid search (pgvector 768-d + ts_rank_cd FTS) with RRF (k=60), 4-gate fail-closed grounding (100% verified, 0% hallucination), and a 2D multi-key Gemini waterfall engine streaming real-time SSE on React 19 / FastAPI.