# harishneel1/vectorless-rag architecture

> Architecture diagram of the GitHub repository harishneel1/vectorless-rag, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-06). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/harishneel1/vectorless-rag
- Repository: https://github.com/harishneel1/vectorless-rag

## Overview

Vectorless RAG is a single-notebook workflow for answering questions over documents without embeddings or a vector store. A user runs the notebook, which loads documents, asks an LLM to identify relevant passages, then uses that context to generate a grounded answer. The README describes these stages, but the notebook source was not provided, so implementation-level calls cannot be verified; edges below reflect documented behavior only.

Main entry points: [`vectorless_rag.ipynb`](https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb) (Passage routing), [`assets`](https://github.com/harishneel1/vectorless-rag/tree/main/assets) (Source documents)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["RAG workflow"]
  node_notebook["Pipeline notebook"]
  node_documents["Source documents"]
  node_load["Document loading"]
  node_route["Passage routing"]
  node_answer["Grounded answer"]
end

node_user(("Notebook user"))
node_llm{{"LLM service"}}

node_user -->|"runs"| node_notebook
node_documents -->|"reads"| node_load
node_notebook -->|"runs pipeline"| node_load
node_load -->|"then routes"| node_route
node_route -->|"prompts for relevance"| node_llm
node_route -->|"passes context"| node_answer
node_answer -->|"generates answer"| node_llm

click node_notebook "https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb"
click node_documents "https://github.com/harishneel1/vectorless-rag/tree/main/assets"
click node_load "https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb"
click node_route "https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb"
click node_answer "https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_notebook,node_documents,node_load,node_route,node_answer,node_user toneBlue
class node_llm toneIndigo
```

## Components

### RAG workflow

- [Pipeline notebook](https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb): `vectorless_rag.ipynb`
- [Source documents](https://github.com/harishneel1/vectorless-rag/tree/main/assets): `assets`
- [Document loading](https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb): `vectorless_rag.ipynb`
- [Passage routing](https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb): `vectorless_rag.ipynb`
- [Grounded answer](https://github.com/harishneel1/vectorless-rag/blob/main/vectorless_rag.ipynb): `vectorless_rag.ipynb`

### Other components

- **Notebook user**
- **LLM service**

## Connections

- Notebook user → Pipeline notebook: runs (evidence: `README.md`)
- Source documents → Document loading: reads (evidence: `README.md`)
- Pipeline notebook → Document loading: runs pipeline (evidence: `README.md`)
- Document loading → Passage routing: then routes (evidence: `README.md`)
- Passage routing → LLM service: prompts for relevance (evidence: `README.md`)
- Passage routing → Grounded answer: passes context (evidence: `README.md`)
- Grounded answer → LLM service: generates answer (evidence: `README.md`)
