Why Vector RAG fails for AI coding agents at scale (And how I used a Neo4j graph to fix it)
Summary
A new open-source tool called Writ uses a hybrid retrieval pipeline with BM25, ONNX vectors, and Neo4j graph traversals to provide context rules for AI coding agents, reducing token bloat by 726x and enforcing plan approval via bash hooks.
Similar Articles
How I wired a Graph DB on top of my vector store to scale 1K agents for 2 months, because vector search alone fails when user preferences change over time.
A detailed architectural guide for building long-running AI agents that handle changing user preferences over time by combining a vector store, graph DB, and temporal edges instead of overwriting data.
@HowToPrompt__: The entire vector database industry just got destroyed by A free tool from 1974. For the last two years, every company …
Researchers report that the classic grep command outperforms modern vector databases in retrieval tasks for autonomous AI agents, challenging the prevailing RAG infrastructure approach.
@pauliusztin_: I spent months optimizing GraphRAG retrieval. But it turned out I was optimizing the wrong thing.... The biggest knowle…
A detailed guide on optimizing knowledge graph ingestion for AI agents, presenting a five-step pipeline (extraction, resolution, embedding, deduplication, routing) to prevent graph corruption and improve retrieval quality.
@_avichawla: 8 RAG architectures for AI Engineers: (explained with usage) 1) Naive RAG - Retrieves documents purely based on vector …
A tweet thread explaining 8 different RAG architectures (Naive, Multimodal, HyDE, Corrective, Graph, Hybrid, Adaptive, Agentic) with their use cases, and hinting at an improved indexing technique.
@techwith_ram: Your AI stack has a database problem You need a vector DB for embeddings. A graph DB for relationships. An application …
HelixDB is a new open-source database built in Rust that combines vector, graph, and other data models into a single engine, backed by Y Combinator. It aims to replace separate vector, graph, and application databases for AI stacks, offering native vector search, graph traversal, and MCP support.