✈️ AeroAgent — Autonomous AOG Quoting Engine

An AI-powered aviation parts quoting pipeline that automates AOG (Aircraft on Ground) email triage, inventory lookup, and professional quote generation using Claude AI.

---

🚀 Live Demo

▶️ Try the Live Pipeline →

---

📸 Preview

Email Input  →  Claude Extracts JSON  →  SQLite Inventory Check  →  AI Quote Generated

Step	What Happens
📧 Email In	Paste any AOG request email
🤖 AI Extraction	Claude parses part number, airline, urgency
🗄️ DB Lookup	Checks inventory + alternate parts
📝 Quote Out	Professional quote email generated instantly

---

🧠 Features

• LLM Email Extraction — Claude parses unstructured AOG emails into structured JSON
• Intelligent Routing — Auto-suggests alternate parts when primary is out of stock
• AI Quote Generation — Professional sales quotes tailored to urgency and customer
• SQLite Inventory DB — 6 real aviation parts with alternates and certifications
• Gradio UI — Clean web interface with sample emails for instant demo

---

🗂️ Project Structure

aeroagent/
├── aeroagent.ipynb          # Main Colab notebook
├── aeroagent.py             # Standalone Python script
├── requirements.txt         # Dependencies
├── .env.example             # Environment variable template
├── .gitignore               # Ignores secrets and cache
├── README.md                # This file
└── .github/
    ├── workflows/
    │   └── ci.yml           # GitHub Actions CI
    └── ISSUE_TEMPLATE/
        ├── bug_report.md
        └── feature_request.md

---

⚙️ Setup & Installation

1. Clone the repo

git clone https://github.com/Milonahmed96/aeroagent.git
cd aeroagent

2. Install dependencies

pip install -r requirements.txt

3. Set your API key

cp .env.example .env
# Edit .env and add your Anthropic API key

4. Run locally

python aeroagent.py

5. Or open in Colab

[

---

🔑 Environment Variables

Variable	Description
ANTHROPIC_API_KEY	Your Anthropic API key — get one at console.anthropic.com

Never commit your API key. Use .env locally or Colab Secrets in Google Colab.

---

🛠️ Tech Stack

Layer	Technology
AI Model	Anthropic Claude Sonnet (claude-sonnet-4-5)
Database	SQLite (via Python sqlite3)
UI	Gradio
Language	Python 3.10+
Hosting	Google Colab / Hugging Face Spaces

---

📦 Sample Parts Database

Part Number	Description	Aircraft	Stock
AV-29341-A	Hydraulic Actuator Flap	Boeing 737-800	3
GE-CF56-7B	CFM56-7B Fan Blade	Boeing 737 NG	0 (alt available)
AIR-A320-BLV	Bleed Air Valve	Airbus A320	5
NGS-777-STR	Nose Gear Steering Actuator	Boeing 777	1

---

🌐 Deployment

Hugging Face Spaces

'https://huggingface.co/spaces/Milon96/aeroagent.py'

Gradio Share Link

demo.launch(share=True)  # Already in the notebook

Copy the 'https://e7bc60d718f33cedcb.gradio.live' URL and share it directly.

---

🗺️ Roadmap

• Add more aircraft types and parts
• Email sending integration (SendGrid)
• PDF quote generation
• Multi-turn conversation memory
• Authentication for multi-user access

---

📄 License

MIT License — see LICENSE

---

👤 Author

Milon Ahmed

Built as part of the AI Engineer Roadmap portfolio.
Connect on LinkedIn | GitHub