# dhinakaran311/job_search architecture

> Architecture diagram of the GitHub repository dhinakaran311/job_search, 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/dhinakaran311/job_search
- Repository: https://github.com/dhinakaran311/job_search

## Overview

This is a Streamlit job-search and recommendation application. A job seeker loads or searches job postings, supplies a resume, and reviews ranked Safe, Stretch, and Fallback matches with skill gaps; optional LLM insights and CSV export support follow-up. The graph preserves the separate ingestion, resume, extraction, retrieval, ranking, and persistence stages. Edges reflect shown calls or documented workflows; some README-described capabilities have unsampled implementation details, so their links are documented rather than inferred from module names.

Main entry points: [`app_streamlit/app.py`](https://github.com/dhinakaran311/job_search/blob/main/app_streamlit/app.py) (Streamlit App), [`backend/services/retrieval.py`](https://github.com/dhinakaran311/job_search/blob/main/backend/services/retrieval.py) (Job Retrieval), [`backend/ingestion/job_loader.py`](https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_loader.py) (Offline Job Loading), [`backend/ingestion/job_api_adzuna.py`](https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_api_adzuna.py) (Adzuna Fetching)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interface["User Interface"]
  node_streamlit["Streamlit App<br/>[app.py]"]
  node_resumeparse["Resume Parsing<br/>[resume_parser.py]"]
  node_export["CSV Export<br/>[app.py]"]
end

subgraph group_ingestion["Job Ingestion"]
  node_offline["Offline Job Loading<br/>[job_loader.py]"]
  node_adzuna["Adzuna Fetching<br/>[job_api_adzuna.py]"]
end

subgraph group_matching["Matching Services"]
  node_skills["Skill Extraction<br/>[skill_extractor.py]"]
  node_llm["LLM Insights<br/>[llm.py]"]
  node_retrieval["Job Retrieval<br/>[retrieval.py]"]
  node_ranking["Scoring and Buckets<br/>[ranking.py]"]
  node_embedding["Text Embeddings<br/>[embedding.py]"]
end

subgraph group_storage["Persistence"]
  node_chroma_helper["Chroma Access<br/>[chroma.py]"]
  node_chroma[("Chroma Store<br/>[chroma.py]")]
end

node_jobseeker(("Job Seeker"))
node_adzuna_service{{"Adzuna API"}}
node_ollama{{"Ollama"}}

node_jobseeker -->|"provides input"| node_streamlit
node_streamlit -.->|"parses resume"| node_resumeparse
node_streamlit -->|"searches jobs"| node_adzuna
node_adzuna -->|"fetches postings"| node_adzuna_service
node_adzuna -->|"delegates ingestion"| node_offline
node_offline -->|"embeds descriptions"| node_embedding
node_offline -->|"gets collection"| node_chroma_helper
node_chroma_helper -->|"creates collection"| node_chroma
node_streamlit -->|"extracts skills"| node_skills
node_skills -->|"requests extraction"| node_llm
node_llm -.->|"generates insights"| node_ollama
node_streamlit -->|"requests recommendations"| node_retrieval
node_retrieval -->|"uses embeddings"| node_embedding
node_retrieval -->|"uses database access"| node_chroma_helper
node_retrieval -->|"uses scoring"| node_ranking
node_retrieval -->|"extracts skills"| node_skills
node_retrieval -.->|"returns matches"| node_streamlit
node_streamlit -->|"requests summaries and plans"| node_llm
node_streamlit -->|"exports results"| node_export

click node_streamlit "https://github.com/dhinakaran311/job_search/blob/main/app_streamlit/app.py"
click node_resumeparse "https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/resume_parser.py"
click node_export "https://github.com/dhinakaran311/job_search/blob/main/app_streamlit/app.py"
click node_offline "https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_loader.py"
click node_adzuna "https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_api_adzuna.py"
click node_skills "https://github.com/dhinakaran311/job_search/blob/main/backend/services/skill_extractor.py"
click node_llm "https://github.com/dhinakaran311/job_search/blob/main/backend/services/llm.py"
click node_retrieval "https://github.com/dhinakaran311/job_search/blob/main/backend/services/retrieval.py"
click node_ranking "https://github.com/dhinakaran311/job_search/blob/main/backend/services/ranking.py"
click node_embedding "https://github.com/dhinakaran311/job_search/blob/main/backend/services/embedding.py"
click node_chroma_helper "https://github.com/dhinakaran311/job_search/blob/main/backend/db/chroma.py"
click node_chroma "https://github.com/dhinakaran311/job_search/blob/main/backend/db/chroma.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_streamlit,node_resumeparse,node_export toneBlue
class node_offline,node_adzuna toneAmber
class node_skills,node_llm,node_retrieval,node_ranking,node_embedding,node_adzuna_service toneMint
class node_chroma_helper,node_chroma toneRose
class node_jobseeker,node_ollama toneIndigo
```

## Components

### User Interface

- [Streamlit App](https://github.com/dhinakaran311/job_search/blob/main/app_streamlit/app.py): `app_streamlit/app.py`
- [Resume Parsing](https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/resume_parser.py): `backend/ingestion/resume_parser.py`
- [CSV Export](https://github.com/dhinakaran311/job_search/blob/main/app_streamlit/app.py): `app_streamlit/app.py`

### Job Ingestion

- [Offline Job Loading](https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_loader.py): `backend/ingestion/job_loader.py`
- [Adzuna Fetching](https://github.com/dhinakaran311/job_search/blob/main/backend/ingestion/job_api_adzuna.py): `backend/ingestion/job_api_adzuna.py`

### Matching Services

- [Skill Extraction](https://github.com/dhinakaran311/job_search/blob/main/backend/services/skill_extractor.py): `backend/services/skill_extractor.py`
- [LLM Insights](https://github.com/dhinakaran311/job_search/blob/main/backend/services/llm.py): `backend/services/llm.py`
- [Job Retrieval](https://github.com/dhinakaran311/job_search/blob/main/backend/services/retrieval.py): `backend/services/retrieval.py`
- [Scoring and Buckets](https://github.com/dhinakaran311/job_search/blob/main/backend/services/ranking.py): `backend/services/ranking.py`
- [Text Embeddings](https://github.com/dhinakaran311/job_search/blob/main/backend/services/embedding.py): `backend/services/embedding.py`

### Persistence

- [Chroma Access](https://github.com/dhinakaran311/job_search/blob/main/backend/db/chroma.py): `backend/db/chroma.py`
- [Chroma Store](https://github.com/dhinakaran311/job_search/blob/main/backend/db/chroma.py): `backend/db/chroma.py`

### Other components

- **Job Seeker**
- **Adzuna API**
- **Ollama**

## Connections

- Job Seeker → Streamlit App: provides input (evidence: `README.md`)
- Streamlit App → Resume Parsing: parses resume (evidence: `README.md`)
- Streamlit App → Adzuna Fetching: searches jobs (evidence: `app_streamlit/app.py`)
- Adzuna Fetching → Adzuna API: fetches postings (evidence: `backend/ingestion/job_api_adzuna.py`)
- Adzuna Fetching → Offline Job Loading: delegates ingestion (evidence: `backend/ingestion/job_api_adzuna.py`)
- Offline Job Loading → Text Embeddings: embeds descriptions (evidence: `backend/ingestion/job_loader.py`)
- Offline Job Loading → Chroma Access: gets collection (evidence: `backend/ingestion/job_loader.py`)
- Chroma Access → Chroma Store: creates collection (evidence: `backend/db/chroma.py`)
- Streamlit App → Skill Extraction: extracts skills (evidence: `app_streamlit/app.py`)
- Skill Extraction → LLM Insights: requests extraction (evidence: `backend/services/skill_extractor.py`)
- LLM Insights → Ollama: generates insights (evidence: `backend/services/llm.py`)
- Streamlit App → Job Retrieval: requests recommendations (evidence: `README.md`)
- Job Retrieval → Text Embeddings: uses embeddings (evidence: `backend/services/retrieval.py`)
- Job Retrieval → Chroma Access: uses database access (evidence: `backend/services/retrieval.py`)
- Job Retrieval → Scoring and Buckets: uses scoring (evidence: `backend/services/retrieval.py`)
- Job Retrieval → Skill Extraction: extracts skills (evidence: `backend/services/retrieval.py`)
- Job Retrieval → Streamlit App: returns matches (evidence: `README.md`)
- Streamlit App → LLM Insights: requests summaries and plans (evidence: `app_streamlit/app.py`)
- Streamlit App → CSV Export: exports results (evidence: `README.md`)
