Welcome to Support Bot
Support Bot is an AI-powered incident resolution platform designed to help teams quickly resolve technical incidents by leveraging LangGraph agents, vector search, and a comprehensive knowledge base. Built with modern technologies, it provides intelligent incident analysis, automated search across historical incidents, and real-time assistance.Key Features
AI Copilot
Intelligent incident resolution assistant powered by LangGraph agents that can search, analyze, and recommend solutions based on historical data and context.
Knowledge Base
Vector-powered search using Qdrant that enables semantic search across incident reports, finding similar issues and proven solutions instantly.
Chat Interface
Interactive chat experience with streaming responses, markdown support, and contextual follow-ups for natural incident investigation.
OAuth & RBAC
Enterprise-grade authentication with OAuth 2.0 support (Google, GitHub, Microsoft) and role-based access control for secure operations.
Multi-LLM Support
Flexible LLM provider support including Anthropic Claude, OpenAI GPT, Google Gemini, and Ollama for local deployments.
Vector Search
Advanced semantic search capabilities using sentence transformers and Qdrant vector database for finding relevant incidents.
What Makes Support Bot Different?
Support Bot combines the power of modern AI frameworks with practical incident management:- Conversational Memory: LangGraph with PostgreSQL checkpointing maintains context across conversations
- Smart Tool Selection: The AI agent automatically selects the right search tool based on your query
- Golden Examples: Learn from verified incident resolutions to provide better recommendations
- Real-time Streaming: Get responses as they’re generated for faster feedback
- Production-Ready: Built with FastAPI, React, and enterprise-grade database systems
Quick Links
Quickstart
Get up and running in minutes with our step-by-step guide
Architecture
Understand the system design and technology stack
API Reference
Explore the REST API endpoints and schemas
Use Cases
Support Bot is ideal for:- DevOps Teams: Quickly find solutions to deployment issues and infrastructure problems
- Customer Support: Retrieve answers from past support tickets and incident reports
- SRE Teams: Analyze patterns in production incidents and automate initial triage
- Knowledge Management: Build a searchable repository of technical solutions
Technology Stack
Support Bot is built with modern, production-ready technologies:- Backend: FastAPI (Python 3.11+) with async support
- Frontend: React 19 with Vite and TailwindCSS
- AI Framework: LangGraph for agentic workflows
- Vector DB: Qdrant for semantic search
- Database: PostgreSQL with pgvector extension
- LLM Providers: Anthropic, OpenAI, Google, Ollama
Support Bot requires Python 3.11 or higher and uses modern async/await patterns throughout the codebase.