# vikasjha2003/askme-ai architecture

> Architecture diagram of the GitHub repository vikasjha2003/askme-ai, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-07). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/vikasjha2003/askme-ai
- Repository: https://github.com/vikasjha2003/askme-ai

## Overview

AskMe AI is a CLI for PDF-grounded question answering and retrieval experiments. In interactive Q&A, a user selects a PDF and chunking method; the app loads and chunks it, creates a Chroma store, then retrieves context and asks a local LLM to answer. Its benchmark mode compares retrieval across chunking strategies against a QA set and can generate per-question answers. The workflow is documented in the README and partly visible in excerpts; benchmark internals are represented only where the README describes them.

Main entry points: [`src/main.py`](https://github.com/vikasjha2003/askme-ai/blob/main/src/main.py) (CLI workflow), [`src/benchmark.py`](https://github.com/vikasjha2003/askme-ai/blob/main/src/benchmark.py) (Retrieval benchmark), [`src/result_generator.py`](https://github.com/vikasjha2003/askme-ai/blob/main/src/result_generator.py) (Answer evaluation), [`src/vector_store.py`](https://github.com/vikasjha2003/askme-ai/blob/main/src/vector_store.py) (Vector indexing)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_ingestion["Document ingestion"]
  node_loader["PDF loader<br/>[doc_loader.py]"]
  node_config["Runtime paths<br/>[config.py]"]
end

subgraph group_retrieval["Chunking and retrieval"]
  node_chunker["Chunk strategies<br/>[chunker.py]"]
  node_embedding["Embeddings<br/>[embedding.py]"]
  node_store_writer["Vector indexing<br/>[vector_store.py]"]
  node_retriever["MMR retrieval<br/>[retriever.py]"]
  node_vector_db[("Chroma store<br/>[vector_store.py]")]
end

subgraph group_answering["Grounded answering"]
  node_rag["RAG prompt<br/>[rag.py]"]
  node_llm["Local generation<br/>[local_llm.py]"]
end

subgraph group_evaluation["Benchmark and results"]
  node_benchmark["Retrieval benchmark<br/>[benchmark.py]"]
  node_qa_set["Evaluation questions"]
  node_result_generator["Answer evaluation"]
  node_results["Saved results"]
end

node_user(("CLI user"))
node_app["CLI workflow<br/>[main.py]"]

node_user -->|"chooses mode"| node_app
node_app -->|"loads PDF"| node_loader
node_app -->|"creates chunks"| node_chunker
node_app -->|"indexes chunks"| node_store_writer
node_app -->|"asks question"| node_rag
node_loader -->|"provides documents"| node_chunker
node_chunker -->|"provides chunks"| node_store_writer
node_store_writer -->|"embeds chunks"| node_embedding
node_store_writer -->|"writes vectors"| node_vector_db
node_retriever -->|"embeds query"| node_embedding
node_retriever -->|"searches vectors"| node_vector_db
node_rag -->|"fetches context"| node_retriever
node_rag -->|"sends prompt"| node_llm
node_app -->|"runs evaluation"| node_benchmark
node_benchmark -->|"reads questions"| node_qa_set
node_benchmark -->|"compares methods"| node_chunker
node_benchmark -->|"builds stores"| node_store_writer
node_benchmark -->|"saves metrics"| node_results
node_app -->|"generates answers"| node_result_generator
node_result_generator -->|"reads questions"| node_qa_set
node_result_generator -->|"retrieves context"| node_retriever
node_result_generator -->|"generates answers"| node_llm
node_result_generator -->|"writes outputs"| node_results
node_config -->|"sets store path"| node_vector_db

click node_app "https://github.com/vikasjha2003/askme-ai/blob/main/src/main.py"
click node_loader "https://github.com/vikasjha2003/askme-ai/blob/main/src/doc_loader.py"
click node_chunker "https://github.com/vikasjha2003/askme-ai/blob/main/src/chunker.py"
click node_embedding "https://github.com/vikasjha2003/askme-ai/blob/main/src/embedding.py"
click node_store_writer "https://github.com/vikasjha2003/askme-ai/blob/main/src/vector_store.py"
click node_retriever "https://github.com/vikasjha2003/askme-ai/blob/main/src/retriever.py"
click node_vector_db "https://github.com/vikasjha2003/askme-ai/blob/main/src/vector_store.py"
click node_rag "https://github.com/vikasjha2003/askme-ai/blob/main/src/rag.py"
click node_llm "https://github.com/vikasjha2003/askme-ai/blob/main/src/local_llm.py"
click node_benchmark "https://github.com/vikasjha2003/askme-ai/blob/main/src/benchmark.py"
click node_qa_set "https://github.com/vikasjha2003/askme-ai/blob/main/src/data/questions_and_answers/apple_2025_annual_report.json"
click node_result_generator "https://github.com/vikasjha2003/askme-ai/blob/main/src/result_generator.py"
click node_config "https://github.com/vikasjha2003/askme-ai/blob/main/src/config.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_loader,node_config,node_user toneBlue
class node_chunker,node_embedding,node_store_writer,node_retriever,node_vector_db toneAmber
class node_rag,node_llm toneMint
class node_benchmark,node_qa_set,node_result_generator,node_results toneRose
class node_app toneTeal
```

## Components

### Document ingestion

- [PDF loader](https://github.com/vikasjha2003/askme-ai/blob/main/src/doc_loader.py): `src/doc_loader.py`
- [Runtime paths](https://github.com/vikasjha2003/askme-ai/blob/main/src/config.py): `src/config.py`

### Chunking and retrieval

- [Chunk strategies](https://github.com/vikasjha2003/askme-ai/blob/main/src/chunker.py): `src/chunker.py`
- [Embeddings](https://github.com/vikasjha2003/askme-ai/blob/main/src/embedding.py): `src/embedding.py`
- [Vector indexing](https://github.com/vikasjha2003/askme-ai/blob/main/src/vector_store.py): `src/vector_store.py`
- [MMR retrieval](https://github.com/vikasjha2003/askme-ai/blob/main/src/retriever.py): `src/retriever.py`
- [Chroma store](https://github.com/vikasjha2003/askme-ai/blob/main/src/vector_store.py): `src/vector_store.py`

### Grounded answering

- [RAG prompt](https://github.com/vikasjha2003/askme-ai/blob/main/src/rag.py): `src/rag.py`
- [Local generation](https://github.com/vikasjha2003/askme-ai/blob/main/src/local_llm.py): `src/local_llm.py`

### Benchmark and results

- [Retrieval benchmark](https://github.com/vikasjha2003/askme-ai/blob/main/src/benchmark.py): `src/benchmark.py`
- [Evaluation questions](https://github.com/vikasjha2003/askme-ai/blob/main/src/data/questions_and_answers/apple_2025_annual_report.json): `src/data/questions_and_answers/apple_2025_annual_report.json`
- [Answer evaluation](https://github.com/vikasjha2003/askme-ai/blob/main/src/result_generator.py): `src/result_generator.py`
- **Saved results**

### Other components

- **CLI user**
- [CLI workflow](https://github.com/vikasjha2003/askme-ai/blob/main/src/main.py): `src/main.py`

## Connections

- CLI user → CLI workflow: chooses mode (evidence: `src/main.py`)
- CLI workflow → PDF loader: loads PDF (evidence: `src/main.py`)
- CLI workflow → Chunk strategies: creates chunks (evidence: `src/main.py`)
- CLI workflow → Vector indexing: indexes chunks (evidence: `src/main.py`)
- CLI workflow → RAG prompt: asks question (evidence: `src/main.py`)
- PDF loader → Chunk strategies: provides documents (evidence: `README.md`)
- Chunk strategies → Vector indexing: provides chunks (evidence: `README.md`)
- Vector indexing → Embeddings: embeds chunks (evidence: `src/vector_store.py`)
- Vector indexing → Chroma store: writes vectors (evidence: `src/vector_store.py`)
- MMR retrieval → Embeddings: embeds query (evidence: `src/retriever.py`)
- MMR retrieval → Chroma store: searches vectors (evidence: `README.md`)
- RAG prompt → MMR retrieval: fetches context (evidence: `src/rag.py`)
- RAG prompt → Local generation: sends prompt (evidence: `src/rag.py`)
- CLI workflow → Retrieval benchmark: runs evaluation (evidence: `src/main.py`)
- Retrieval benchmark → Evaluation questions: reads questions (evidence: `README.md`)
- Retrieval benchmark → Chunk strategies: compares methods (evidence: `README.md`)
- Retrieval benchmark → Vector indexing: builds stores (evidence: `README.md`)
- Retrieval benchmark → Saved results: saves metrics (evidence: `README.md`)
- CLI workflow → Answer evaluation: generates answers (evidence: `src/main.py`)
- Answer evaluation → Evaluation questions: reads questions (evidence: `README.md`)
- Answer evaluation → MMR retrieval: retrieves context (evidence: `src/result_generator.py`)
- Answer evaluation → Local generation: generates answers (evidence: `src/result_generator.py`)
- Answer evaluation → Saved results: writes outputs
- Runtime paths → Chroma store: sets store path (evidence: `src/vector_store.py`)
