A developer advocates for self-hosting ArangoDB, a multi-model database that natively supports graph, document, and vector search with a single query language (AQL), eliminating the need to duct-tape …
Voice agent latency degrades after turn 7-8 in production, occasionally crossing Retell's ~3s reconnect threshold and dropping calls, despite using keep_alive: -1, KV cache reuse, and two-WebSocket ra…
A new open-source RAG agent combines Qdrant hybrid search (dense + BM25 with RRF fusion), LangGraph ReAct routing with safety overrides, an MCP server for tool exposure, and Arize Phoenix for observab…
A developer shipped trelix v2.7.0 with a release pipeline bug that caused binary assets to collide, making it impossible to tell whether the surviving binary was macOS or Linux. Over six subsequent re…
A developer compares Qdrant and Pinecone for production RAG systems, highlighting that Qdrant is an open-source vector search engine that can be self-hosted while Pinecone is a fully managed cloud ser…
A controlled experiment comparing hybrid retrieval pipelines using BM25 versus miniCOIL on the MedQuAD dataset with EmbeddingGemma and Qdrant shows that miniCOIL, a sparse neural retrieval model, impr…
A developer's practical guide details the engineering challenges and solutions for deploying production-ready Retrieval-Augmented Generation (RAG) applications, covering data indexing, vector stores, …
Real-time voice agents using local LLMs face a hard latency ceiling of ~3 seconds imposed by Retell AI's WebSocket timeout, which forces reconnections and call termination if the first token is not pr…
A new open-source project demonstrates how multimodal vector search works under the hood, using OpenAI CLIP (ViT-B/32) to embed text and images into a 512-dimensional vector space and Qdrant (HNSW ind…
A developer built RE-call, a retrieval-augmented memory system for AI agents that runs entirely on PostgreSQL, using pgvector for dense vector search and built-in full-text search, fused via Reciproca…
A developer at an AI startup reduced memory state inconsistencies by 90% by replacing manual testing with a Pytest-based automated test suite for LLM memory stores. The new approach verifies consisten…
Eugene v0.8 introduces memory for AI agents built in Rust, shipping two complementary stores, a VectorStore trait, and agentic RAG through new skills 'remember' and 'recall'. The memory system uses ma…
A developer built a hybrid RAG API that combines dense and sparse retrieval with reciprocal rank fusion (RRF) to answer questions from PDFs. The system uses Qdrant for vector storage, a cross-encoder …
HNSW (Hierarchical Navigable Small World) is a graph-based algorithm that powers vector search in FAISS, pgvector, Qdrant, Weaviate, and Milvus, enabling approximate nearest neighbor search in millise…
A developer built a modular RAG pipeline architecture for production, using Docker containers on Cloud Run, with external configuration to swap components. The pipeline uses 500-character chunks with …
A developer built myRAG, a fully self-hosted RAG stack combining FastAPI, React, Qdrant, PostgreSQL, and Neo4j. The system uses hybrid search with dense and sparse embeddings, cross-encoder reranking,…
A developer built a personal health knowledge base using a vector database and RAG pipeline to organize scattered medical records. The system uses Qdrant for similarity search, Unstructured.io for par…
A developer compares Mem0 and TurboMem as memory layers for TypeScript AI agents. TurboMem runs as a native TypeScript library inside the process, avoiding the separate service and infrastructure requ…
Qdrant reduced retrieval-augmented generation token costs by 67% using native ColBERT reranking, which performs token-to-token matrix comparison inside the database in a single query call, eliminating…
A developer explains how vector search works under the hood, focusing on the two dominant algorithms: IVF (Inverted File Index) and HNSW (Hierarchical Navigable Small World). The post details why appr…