# sonphattrandeveloper/it-job-posting architecture

> Architecture diagram of the GitHub repository sonphattrandeveloper/it-job-posting, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-07-19). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/sonphattrandeveloper/it-job-posting
- Repository: https://github.com/sonphattrandeveloper/it-job-posting

## Overview

##### Project type and purpose
- This is a data analysis and visualization repository built around a single CSV dataset of 3,101 Vietnamese IT job postings.
- Its primary purpose is to store, process, enrich (via LLM‐based extraction of languages/technologies/cities), and render interactive visualizations in a standalone HTML file.

##### File structure
- Top‐level files only:
- job_descriptions.csv – the raw dataset
- data_analysis_visualization.html – the rendered interactive dashboard
- README.md – project description and data schema
- No “src” or “scripts” directories are present; processing is implied to happen offline (e.g., in notebooks) before generating the HTML.

##### README for additional insights
- The CSV has columns for title, company, location, raw description, plus fields extracted via LLMs (main_programming_languages, key_technologies, city).

##### Main components and their interactions
a. Data Storage
b. Data Processing / ETL
- Internals:
– CSV Reader (e.g., pandas)
– LLM Extraction module (e.g., OpenAI API client)
– Cleaner & Transformer
- Outputs a JSON/JS data blob for visualization
c. Visualization Engine
- Consumes the JSON/JS data blob and renders charts, maps, tables
d. End User

Relationships / Data Flow

##### Architectural patterns and design principles
- ETL Pipeline Pattern: clear separation of Ingestion, Enrichment, and Presentation layers
- Single‐page static visualization: front‐end only, no backend server at runtime

##### Relevant technologies and libraries
- Data processing: Python (pandas), OpenAI SDK (or other LLM client)
- Visualization: plain HTML, JavaScript libraries (D3.js, Plotly.js, Chart.js, etc.)
- Deployment: serves as a static HTML file—no container or server framework required

## Diagram (Mermaid)

```mermaid
flowchart TB
    subgraph "Data Layer"
        DS["job_descriptions.csv"]:::data
        README["README.md"]:::data
    end

    subgraph "Processing Layer"
        subgraph "ETL & Enrichment"
            CSVReader["CSV Reader (pandas)"]:::process
            Cleaner["Cleaner & Transformer"]:::process
            LLMExt["LLM Extraction Module"]:::process
        end
    end

    LLMAPI["LLM API Service"]:::external

    subgraph "Presentation Layer"
        Viz["Interactive Dashboard (HTML/JS/D3, Plotly)"]:::presentation
    end

    User["Browser / End User"]:::user

    DS -->|"CSV read"| CSVReader
    CSVReader -->|"clean data"| Cleaner
    Cleaner -->|"text for enrichment"| LLMExt
    LLMExt -->|"JSON blob"| Viz
    Viz -->|"Local open / HTTP GET"| User
    LLMExt -.->|"API call"| LLMAPI

    click DS "https://github.com/sonphattrandeveloper/it-job-posting/blob/main/job_descriptions.csv"
    click README "https://github.com/sonphattrandeveloper/it-job-posting/blob/main/README.md"
    click Viz "https://github.com/sonphattrandeveloper/it-job-posting/blob/main/data_analysis_visualization.html"

    classDef data fill:#cce5ff,stroke:#00529B,stroke-width:2px
    classDef process fill:#d5f5e3,stroke:#239B56,stroke-width:2px
    classDef presentation fill:#fdedcc,stroke:#b36b00,stroke-width:2px
    classDef external fill:#f0f0f0,stroke:#7f8c8d,stroke-width:2px,stroke-dasharray:5 5
    classDef user fill:#e8daef,stroke:#6c3483,stroke-width:2px
```
