# saicharan67778/queriom-ai architecture

> Architecture diagram of the GitHub repository saicharan67778/queriom-ai, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-16). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/saicharan67778/queriom-ai
- Repository: https://github.com/saicharan67778/queriom-ai

## Overview

Queriom is a Streamlit document-intelligence app: users upload PDFs or TXT files, ask grounded questions, or request executive reports. Ingestion extracts page-aware text and chunks it; indexing enables dense and lexical retrieval, while report and Q&A synthesis use Groq and provide citations. Feedback is logged and approved or corrected answers can enter semantic memory; the database also supports DPO export. The README documents additional capabilities, but some UI wiring and implementation details fall outside the excerpts, so uncertain links are omitted.

Main entry points: [`app.py`](https://github.com/saicharan67778/queriom-ai/blob/main/app.py) (Streamlit App), [`core/feedback_db.py`](https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py) (Feedback Database), [`core/rag_pipeline.py`](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py) (Q&A Synthesis), [`core/document_processor.py`](https://github.com/saicharan67778/queriom-ai/blob/main/core/document_processor.py) (Document Processor)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_ui["User Interface"]
  node_app["Streamlit App<br/>[app.py]"]
end

subgraph group_ingest["Document Ingestion"]
  node_processor["Document Processor"]
end

subgraph group_rag["Retrieval and Synthesis"]
  node_index[("Corpus Index<br/>[rag_pipeline.py]")]
  node_vector[("Dense Retrieval<br/>[rag_pipeline.py]")]
  node_bm25["BM25 Retrieval<br/>[rag_pipeline.py]"]
  node_fusion["Hybrid Retrieval<br/>[rag_pipeline.py]"]
  node_gate["Distance Gate<br/>[rag_pipeline.py]"]
  node_qa["Q&amp;A Synthesis<br/>[rag_pipeline.py]"]
  node_reports["Report Synthesis<br/>[rag_pipeline.py]"]
  node_digest["Corpus Digest<br/>[rag_pipeline.py]"]
end

subgraph group_memory["Feedback and Memory"]
  node_feedback[("Feedback Database<br/>[feedback_db.py]")]
  node_logs[("Query Logs<br/>[feedback_db.py]")]
  node_memory_store[("Semantic Memory<br/>[feedback_db.py]")]
  node_dpo["DPO Export<br/>[feedback_db.py]"]
end

node_user(("User"))
node_groq["Groq Inference"]

node_user -->|"uploads and asks"| node_app
node_app -->|"processes uploads"| node_processor
node_app -->|"initializes index"| node_index
node_processor -->|"supplies chunks"| node_index
node_app -->|"requests digest"| node_digest
node_app -->|"submits questions"| node_qa
node_app -->|"requests reports"| node_reports
node_index -->|"adds documents"| node_vector
node_index -->|"builds retriever"| node_bm25
node_qa -->|"searches similarity"| node_vector
node_qa -->|"retrieves terms"| node_bm25
node_reports -->|"searches context"| node_vector
node_reports -.->|"uses hybrid search"| node_bm25
node_vector -->|"supplies matches"| node_fusion
node_bm25 -->|"supplies matches"| node_fusion
node_fusion -->|"checks relevance"| node_gate
node_gate -->|"passes context"| node_qa
node_qa -->|"generates answer"| node_groq
node_reports -->|"generates report"| node_groq
node_digest -->|"summarizes corpus"| node_groq
node_app -->|"records feedback"| node_feedback
node_feedback -->|"logs queries and edits"| node_logs
node_feedback -->|"indexes approved answers"| node_memory_store
node_memory_store -->|"returns similar queries"| node_feedback
node_feedback -->|"exports preference pairs"| node_dpo
node_app -.->|"looks up memory"| node_feedback
node_feedback -.->|"provides exemplars"| node_qa

click node_app "https://github.com/saicharan67778/queriom-ai/blob/main/app.py"
click node_processor "https://github.com/saicharan67778/queriom-ai/blob/main/core/document_processor.py"
click node_index "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_vector "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_bm25 "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_fusion "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_gate "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_qa "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_reports "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_digest "https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py"
click node_feedback "https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py"
click node_logs "https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py"
click node_memory_store "https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py"
click node_dpo "https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py"

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_app,node_user toneBlue
class node_processor toneAmber
class node_index,node_vector,node_bm25,node_fusion,node_gate,node_qa,node_reports,node_digest toneMint
class node_feedback,node_logs,node_memory_store,node_dpo toneRose
class node_groq toneIndigo
```

## Components

### User Interface

- [Streamlit App](https://github.com/saicharan67778/queriom-ai/blob/main/app.py): `app.py`

### Document Ingestion

- [Document Processor](https://github.com/saicharan67778/queriom-ai/blob/main/core/document_processor.py): `core/document_processor.py`

### Retrieval and Synthesis

- [Corpus Index](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Dense Retrieval](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [BM25 Retrieval](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Hybrid Retrieval](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Distance Gate](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Q&A Synthesis](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Report Synthesis](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`
- [Corpus Digest](https://github.com/saicharan67778/queriom-ai/blob/main/core/rag_pipeline.py): `core/rag_pipeline.py`

### Feedback and Memory

- [Feedback Database](https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py): `core/feedback_db.py`
- [Query Logs](https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py): `core/feedback_db.py`
- [Semantic Memory](https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py): `core/feedback_db.py`
- [DPO Export](https://github.com/saicharan67778/queriom-ai/blob/main/core/feedback_db.py): `core/feedback_db.py`

### Other components

- **User**
- **Groq Inference**

## Connections

- User → Streamlit App: uploads and asks (evidence: `README.md`)
- Streamlit App → Document Processor: processes uploads (evidence: `app.py`)
- Streamlit App → Corpus Index: initializes index (evidence: `app.py`)
- Document Processor → Corpus Index: supplies chunks (evidence: `app.py`)
- Streamlit App → Corpus Digest: requests digest (evidence: `app.py`)
- Streamlit App → Q&A Synthesis: submits questions (evidence: `app.py`)
- Streamlit App → Report Synthesis: requests reports (evidence: `app.py`)
- Corpus Index → Dense Retrieval: adds documents (evidence: `core/rag_pipeline.py`)
- Corpus Index → BM25 Retrieval: builds retriever (evidence: `core/rag_pipeline.py`)
- Q&A Synthesis → Dense Retrieval: searches similarity (evidence: `core/rag_pipeline.py`)
- Q&A Synthesis → BM25 Retrieval: retrieves terms (evidence: `core/rag_pipeline.py`)
- Report Synthesis → Dense Retrieval: searches context (evidence: `README.md`)
- Report Synthesis → BM25 Retrieval: uses hybrid search (evidence: `core/rag_pipeline.py`)
- Dense Retrieval → Hybrid Retrieval: supplies matches (evidence: `core/rag_pipeline.py`)
- BM25 Retrieval → Hybrid Retrieval: supplies matches (evidence: `core/rag_pipeline.py`)
- Hybrid Retrieval → Distance Gate: checks relevance (evidence: `core/rag_pipeline.py`)
- Distance Gate → Q&A Synthesis: passes context (evidence: `core/rag_pipeline.py`)
- Q&A Synthesis → Groq Inference: generates answer (evidence: `core/rag_pipeline.py`)
- Report Synthesis → Groq Inference: generates report (evidence: `core/rag_pipeline.py`)
- Corpus Digest → Groq Inference: summarizes corpus (evidence: `core/rag_pipeline.py`)
- Streamlit App → Feedback Database: records feedback (evidence: `app.py`)
- Feedback Database → Query Logs: logs queries and edits (evidence: `core/feedback_db.py`)
- Feedback Database → Semantic Memory: indexes approved answers (evidence: `core/feedback_db.py`)
- Semantic Memory → Feedback Database: returns similar queries (evidence: `core/feedback_db.py`)
- Feedback Database → DPO Export: exports preference pairs (evidence: `core/feedback_db.py`)
- Streamlit App → Feedback Database: looks up memory (evidence: `app.py`)
- Feedback Database → Q&A Synthesis: provides exemplars
