NextStep - Modular Email Classification Pipeline

Overview

NextStep is a modular email classification pipeline built in Python. It analyzes recruitment-related emails and categorizes them into:

• Moving Forward
• Rejection
• Neutral

The project is designed following clean architecture principles and structured as a scalable data-processing system.

---

Architecture

The project follows separation of concerns:

• 'classifier.py' → Core business logic (signal detection & status decision)
• 'persistence.py' → Data storage layer (JSON handling, ID generation, timestamps)
• 'main.py' → Orchestration layer
• 'results.json' → Structured output storage

This modular design allows future expansion (database, APIs, analytics).

---

Currect Features

• Keyword-based email classification
• Batch email processing
• Structured JSON persistance
• Unique ID generation (UUID4)
• ISO 8601 timestamp tracking (timezone-aware)
• Corrupted JSON handling
• Modular architecture (clean separation of responsabilities)

---

Example Output Structure

Each processed email is stored as:

{
  "id": "uuid",
  "timestamp": "ISO8601",
  "email_text": "original email",
  "status": "Moving forward / Rejection / Neutral",
  "signals": {
      "rejection": [],
      "moving_forward": [],
      "neutral": []
  }
}

---

Tech Stack

• Python 3.10+
• JSON
• Standard Library (uuid, datetime, pathlib)

---

Roadmap

• Regex-based word boundary matching
• Pandas analytics layer
• CSV export
• Database integration
• Gmail API ingestion

---

Author

Abril Rocío Martínez Law Student | Data Engineering Apprenticegit push