OpsMind AI — Multi-Agent Incident RCA Architecture

AI-powered incident root cause analysis platform for DevOps and SRE teams.

Problem Statement

During outages, engineers waste valuable time searching logs, dashboards, and alerts to identify the root cause.

Solution: An AI agent that connects with monitoring tools like Datadog, Grafana, or New Relic, analyzes logs and incidents in real-time, identifies probable root causes, and suggests fixes instantly.

Features

• Multi-agent workflow orchestration using LangGraph
• Retrieval-Augmented Generation (RAG) for historical incident matching
• FAISS vector similarity search
• Monitoring platform connector architecture
• Automated incident timeline generation
• Impacted service detection
• Dynamic incident metrics visualization
• AI system evaluation dashboard
• Downloadable incident reports
• Streamlit-based observability dashboard

Architecture

Tech Stack

• Python
• Streamlit
• LangGraph
• FAISS
• Groq LLM API
• SentenceTransformers

Installation

1. Clone the Repository

git clone https://github.com/Anucool419/OpsMind-AI.git

cd OpsMind-AI

2. Create Virtual Environment

python -m venv venv

Activate environment:

Windows

venv\Scripts\activate

Mac/Linux

source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Configure Environment Variables

Create a .env file:

GROQ_API_KEY=your_api_key

5. Run the Application

streamlit run app/streamlit_app.py

Screenshots

Demo

Video link : https://www.youtube.com/watch?v=OTj5cE5ortQ

Deployed link : https://opsmind-ai-fuonkmwfprksqhivxcddh6.streamlit.app/

Evaluation Metrics

OpsMind AI includes an evaluation layer to measure system reliability and incident analysis quality.

Metrics Tracked

Metric	Description
Retrieval Accuracy	Measures whether relevant historical incidents were retrieved correctly
RCA Match Accuracy	Measures similarity between generated RCA and expected RCA
Severity Accuracy	Evaluates incident severity classification correctness
Average Latency	Measures end-to-end AI analysis response time
Correlation Confidence	Indicates confidence in incident correlation analysis

Future Improvements

• Real-time observability ingestion
• Slack/MS Teams alert integrations
• Kubernetes event streaming
• Live Datadog/New Relic APIs
• Autonomous remediation agents
• Multi-tenant incident analytics

Note

This project uses simulated observability logs and monitoring connectors to demonstrate incident analysis workflows in a production-inspired environment. The architecture is designed to support integration with real monitoring platforms such as Datadog, Grafana, and New Relic APIs.

Contributors

• Ananya Srinivasan
  • AI Agent Workflow
  • RAG + FAISS Retrieval
  • LangGraph Orchestration
  • Streamlit Dashboard
  • Evaluation Framework