# udayallu/rag-multi-corpus architecture

> Architecture diagram of the GitHub repository udayallu/rag-multi-corpus, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-26). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/udayallu/rag-multi-corpus
- Repository: https://github.com/udayallu/rag-multi-corpus

## Overview

RAG-Multi-Corpus is a synthetic, multi-domain document dataset and evaluation benchmark, not a RAG application. Its core use is to supply enterprise corpora in multiple formats and a curated question set for evaluating retrieval and answer quality. The README documents five organizations and benchmark use; no runtime code or transformation pipeline is present in the supplied tree, so the graph represents dataset assets and documented evaluation relationships only.

Main entry points: [`datasets/Aventro Motors`](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Aventro%20Motors) (Aventro Motors), [`datasets/CloudWay-24`](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/CloudWay-24) (CloudWay 24), [`datasets/Cendara University`](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Cendara%20University) (Cendara University), [`datasets/Velvera Technologies`](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Velvera%20Technologies) (Velvera Technologies)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_corpora["Enterprise corpora"]
  node_avent[("Aventro Motors")]
  node_cloudway[("CloudWay 24")]
  node_cendara[("Cendara University")]
  node_velvera[("Velvera Technologies")]
  node_zxbank[("ZX Bank")]
end

subgraph group_formats["Document formats"]
  node_pdf["PDF documents"]
  node_html["HTML documents"]
  node_docx["DOCX documents"]
  node_pptx["PPTX documents"]
  node_markdown["Markdown documents"]
end

subgraph group_benchmark["Benchmark assets"]
  node_qa["Curated Q&A"]
  node_query_categories["Query categories"]
  node_chunks["Aggregated chunks"]
  node_parsed_chunks["Parsed chunks"]
  node_lean_chunks["Reduced-token chunks"]
end

node_rag_evaluator(("RAG system"))

node_avent -->|"available as"| node_pdf
node_avent -->|"available as"| node_html
node_avent -->|"available as"| node_docx
node_avent -->|"available as"| node_pptx
node_avent -->|"available as"| node_markdown
node_cloudway -->|"available as"| node_pdf
node_cloudway -->|"available as"| node_html
node_cloudway -->|"available as"| node_docx
node_cloudway -->|"available as"| node_pptx
node_cloudway -->|"available as"| node_markdown
node_cendara -->|"available as"| node_pdf
node_cendara -->|"available as"| node_html
node_cendara -->|"available as"| node_docx
node_cendara -->|"available as"| node_pptx
node_cendara -->|"available as"| node_markdown
node_velvera -->|"available as"| node_pdf
node_velvera -->|"available as"| node_html
node_velvera -->|"available as"| node_docx
node_velvera -->|"available as"| node_pptx
node_velvera -->|"available as"| node_markdown
node_zxbank -->|"available as"| node_pdf
node_zxbank -->|"available as"| node_html
node_zxbank -->|"available as"| node_docx
node_zxbank -->|"available as"| node_pptx
node_zxbank -->|"available as"| node_markdown
node_rag_evaluator -->|"evaluates against"| node_avent
node_rag_evaluator -->|"evaluates against"| node_cloudway
node_rag_evaluator -->|"evaluates against"| node_cendara
node_rag_evaluator -->|"evaluates against"| node_velvera
node_rag_evaluator -->|"evaluates against"| node_zxbank
node_qa -->|"tests answers"| node_rag_evaluator
node_query_categories -->|"classifies queries"| node_qa
node_chunks -->|"supports benchmarking"| node_rag_evaluator
node_parsed_chunks -->|"supports benchmarking"| node_rag_evaluator
node_lean_chunks -->|"supports benchmarking"| node_rag_evaluator

click node_avent "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Aventro%20Motors"
click node_cloudway "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/CloudWay-24"
click node_cendara "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Cendara%20University"
click node_velvera "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Velvera%20Technologies"
click node_zxbank "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/ZX%20Bank"
click node_pdf "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets"
click node_html "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets"
click node_docx "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets"
click node_pptx "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets"
click node_markdown "https://github.com/udayallu/rag-multi-corpus/tree/main/datasets"
click node_qa "https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references"
click node_query_categories "https://github.com/udayallu/rag-multi-corpus/blob/main/bechmark/bechmark-agentic-references/Dataset%20categories%20-%20queries.csv"
click node_chunks "https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references"
click node_parsed_chunks "https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references/parsed-chunks"
click node_lean_chunks "https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references/parsed-chunks-less-tokens"

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_avent,node_cloudway,node_cendara,node_velvera,node_zxbank toneBlue
class node_pdf,node_html,node_docx,node_pptx,node_markdown toneAmber
class node_qa,node_query_categories,node_chunks,node_parsed_chunks,node_lean_chunks toneMint
class node_rag_evaluator toneIndigo
```

## Components

### Enterprise corpora

- [Aventro Motors](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Aventro%20Motors): `datasets/Aventro Motors`
- [CloudWay 24](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/CloudWay-24): `datasets/CloudWay-24`
- [Cendara University](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Cendara%20University): `datasets/Cendara University`
- [Velvera Technologies](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/Velvera%20Technologies): `datasets/Velvera Technologies`
- [ZX Bank](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets/ZX%20Bank): `datasets/ZX Bank`

### Document formats

- [PDF documents](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets): `datasets`
- [HTML documents](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets): `datasets`
- [DOCX documents](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets): `datasets`
- [PPTX documents](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets): `datasets`
- [Markdown documents](https://github.com/udayallu/rag-multi-corpus/tree/main/datasets): `datasets`

### Benchmark assets

- [Curated Q&A](https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references): `bechmark/bechmark-agentic-references`
- [Query categories](https://github.com/udayallu/rag-multi-corpus/blob/main/bechmark/bechmark-agentic-references/Dataset%20categories%20-%20queries.csv): `bechmark/bechmark-agentic-references/Dataset categories - queries.csv`
- [Aggregated chunks](https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references): `bechmark/bechmark-agentic-references`
- [Parsed chunks](https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references/parsed-chunks): `bechmark/bechmark-agentic-references/parsed-chunks`
- [Reduced-token chunks](https://github.com/udayallu/rag-multi-corpus/tree/main/bechmark/bechmark-agentic-references/parsed-chunks-less-tokens): `bechmark/bechmark-agentic-references/parsed-chunks-less-tokens`

### Other components

- **RAG system**

## Connections

- Aventro Motors → PDF documents: available as
- Aventro Motors → HTML documents: available as
- Aventro Motors → DOCX documents: available as
- Aventro Motors → PPTX documents: available as
- Aventro Motors → Markdown documents: available as
- CloudWay 24 → PDF documents: available as
- CloudWay 24 → HTML documents: available as
- CloudWay 24 → DOCX documents: available as
- CloudWay 24 → PPTX documents: available as
- CloudWay 24 → Markdown documents: available as
- Cendara University → PDF documents: available as
- Cendara University → HTML documents: available as
- Cendara University → DOCX documents: available as
- Cendara University → PPTX documents: available as
- Cendara University → Markdown documents: available as
- Velvera Technologies → PDF documents: available as
- Velvera Technologies → HTML documents: available as
- Velvera Technologies → DOCX documents: available as
- Velvera Technologies → PPTX documents: available as
- Velvera Technologies → Markdown documents: available as
- ZX Bank → PDF documents: available as
- ZX Bank → HTML documents: available as
- ZX Bank → DOCX documents: available as
- ZX Bank → PPTX documents: available as
- ZX Bank → Markdown documents: available as
- RAG system → Aventro Motors: evaluates against
- RAG system → CloudWay 24: evaluates against
- RAG system → Cendara University: evaluates against
- RAG system → Velvera Technologies: evaluates against
- RAG system → ZX Bank: evaluates against
- Curated Q&A → RAG system: tests answers
- Query categories → Curated Q&A: classifies queries
- Aggregated chunks → RAG system: supports benchmarking
- Parsed chunks → RAG system: supports benchmarking
- Reduced-token chunks → RAG system: supports benchmarking