syed7741/aegis-industrial-ai-dataset

AEGIS Industrial AI Dataset

A synthetic industrial operations dataset created for AEGIS AI, an end-to-end industrial artificial intelligence platform.

The dataset is designed to support experimentation with:

• —Retrieval-Augmented Generation (RAG)
• —Semantic Search
• —Industrial AI Assistants
• —Document Intelligence
• —Predictive Maintenance
• —Worker Safety
• —Robot Monitoring
• —Computer Vision / Inspection
• —AI Alerts
• —Workflow Automation

Project Overview

AEGIS AI is an industrial intelligence platform designed to demonstrate how modern AI systems can combine operational data, semantic retrieval, local language models, computer vision, and workflow automation.

The project is being developed as an end-to-end AI engineering portfolio project using:

• —Python
• —FastAPI
• —React
• —TypeScript
• —Hugging Face Datasets
• —Sentence Transformers
• —Vector Search
• —Retrieval-Augmented Generation
• —Local Language Models
• —PostgreSQL
• —Computer Vision
• —AI Workflow Automation

The objective is to demonstrate the architecture of a real AI application rather than only an isolated machine-learning notebook.

Dataset Structure

The dataset currently contains 64 synthetic industrial records covering multiple operational domains.

Each record contains structured metadata together with natural-language operational knowledge that can be indexed and retrieved by an AI system.

Example fields include:

AEGIS Modules

The dataset includes records from eight major AEGIS AI modules.

Worker Safety

Industrial safety procedures, unsafe operating conditions, lockout/tagout scenarios, and recommended safety responses.

Predictive Maintenance

Equipment degradation, maintenance observations, abnormal operational behavior, and recommended maintenance actions.

Robot Monitoring

Industrial robot events, robot status information, operating abnormalities, and robot-related alerts.

Vision Inspection

Synthetic visual inspection findings representing manufacturing or industrial quality-control events.

AI Alerts

Machine and operational alerts designed for automated AI prioritization.

Workflow Automation

Operational events that can trigger automated industrial workflows.

Document Assistant

Knowledge records designed for retrieval, semantic search, and AI-assisted question answering.

Factory Status

Operational status information representing industrial facilities and equipment.

Industries Represented

The synthetic records cover multiple industrial environments including:

• —Manufacturing
• —Oil & Gas
• —Warehousing / Logistics
• —Robotics

Using the Dataset

Install Hugging Face Datasets:

pip install datasets