(Let's start with a) Scalable question-answering system utilizing FastAPI, LangChain (LCEL), and PGVector, featuring an ingestion pipeline. Deployed on GCP Cloud Run via Terraform.

Knowledge base Q&A program using LangChain for retrieval-augmented prompting and PGVector as vector store.

docker compose setup w/ postgis, pgvector, and govgis_nov2023-slim-spatial

いにしえの人工無能（人工無脳）っぽいチャット

Generative AI demos, an example of a langchain based application that implements Retrieval-Augmented-Generation for an enhanced generation. It uses OpenAI model and OpenAI function tool agent. Able to run Ollama Mistral local LLM too.

Implement RAG (using LangChain and PostgreSQL) for Go applications to improve the accuracy and relevance of LLM outputs

smarter search assistant for pascal college norms 🧠

PrivateQ is a RAG-based LLM application which integrates private knowledge library to improve response relevance and query accuracy

Easily batch create vectors with OpenAI

A search application using Aurora Postgresql and pgvector for an online retail store product catalog

a langchain prototype using pgvector vector store and llama2 embeddings

Local deployment of different vector databases using Minikube

Load Terms of Service and Privacy Policies into vectordb, semantic search, and question answering.

Lab environment for demostrate building a simple search engine with LLM model and PostgreSQL database with pgvector extension

Image Vector Similarity Search with Azure AI Vision (Florence model) and Azure Cosmos DB for PostgreSQL

Example of a Retrieval-Augmented Generation with Postgres, pgvector, ollama, Llama3 and Go.

CLI Voice Assistant GPT (pgvector, AWS polly, whisper)

OCR and Automatically Categorize Documents, Full-Stack Project

VLib is a digital library platform targeting college library systems that utilises a vector database for discovering required resources and thereby making information accessible to all users irrespective of their knowledge level. It overcomes the incapability of present systems to handle descriptive queries thereby limiting information access.

Semantic Search by embedding and vector db