☁️ CloudOps AI Assistant

An intelligent, context-aware chatbot designed to help DevOps and MLOps engineers troubleshoot cloud infrastructure, CI/CD pipelines, and ML deployments with speed and accuracy.

Built for the NeoStats AI Engineer Use Case.

🚀 Live Demo

CloudOps AI Assistant - Streamlit App

---

✨ Features

• 🧠 Retrieval-Augmented Generation (RAG): Integrates a local knowledge base (AWS, Docker, MLOps troubleshooting) using FAISS and Sentence Transformers.
• 🌐 Real-time Web Search: Leverages DuckDuckGo Search to fetch the latest fixes and documentation when local knowledge is insufficient.
• ⚡ High-Speed Inference: Powered by Groq using the latest Llama-3.3-70b-versatile model.
• Toggleable Modes:
  • Concise: Quick, bulleted summaries for fast resolution.
  • Detailed: In-depth explanations with CLI commands and best practices.
• Bonus UI Features:
  • Suggested Questions: One-click common queries (S3 errors, Docker loops, etc.).
  • Source Transparency: Expandable sections showing exactly what local and web sources were analyzed.

---

🛠️ Tech Stack

• UI Framework: Streamlit
• LLM API: Groq (Llama-3.3-70b-versatile)
• Vector DB: FAISS (Facebook AI Similarity Search)
• Embeddings: Sentence-Transformers (all-MiniLM-L6-v2)
• Search API: DuckDuckGo Search (DDGS)
• Language: Python 3.10+

---

🔧 Local Setup

1. Clone the repository:

   git clone https://github.com/singhharsh77/cloudops-ai-assistant.git
   cd cloudops-ai-assistant

2. Install dependencies:

   pip install -r requirements.txt

3. Configure API Key: Create a .env file in the root directory:

   GROQ_API_KEY=your_groq_api_key_here

4. Run the application:

   streamlit run app.py

---

📁 Project Structure

project/
├── config/
│   └── config.py          # API key & environment management
├── data/
│   └── docs/              # Knowledge base for RAG (AWS, Docker, MLOps)
├── models/
│   ├── llm.py             # Groq SDK integration
│   └── embeddings.py      # Sentence Transformers logic
├── utils/
│   ├── rag.py             # FAISS indexing and retrieval logic
│   └── web_search.py      # Real-time search implementation
├── app.py                 # Main Streamlit UI & logic
├── requirements.txt       # Project dependencies
└── README.md              # Project documentation

---

📝 Troubleshooting & Assumptions

• Embeddings: Used a lightweight local model (all-MiniLM-L6-v2) to ensure fast performance on Streamlit Cloud without heavy GPU requirements.
• Reliability: All functions are wrapped in robust try-except blocks to handle API outages or missing documents gracefully.

---

👨‍💻 Author

Harsh Singh
AWS Certified Solutions Architect – Associate
AI & Cloud - DevOps Enthusiast