This AI-native customer service platform, Tentix, addresses inefficiencies in support resolution by offering accelerated response times and reduced human intervention. Targeting businesses seeking to enhance customer satisfaction and operational efficiency, Tentix leverages AI to automate and streamline support workflows, aiming for a 10x improvement in key service metrics.

How It Works

Tentix employs a modular architecture centered around an AI agent orchestrated by LangGraph. Ticket data, general documentation, and starred conversations are processed to build a Unified Vector Knowledge Base, primarily using PostgreSQL with the pgvector extension. This KB serves as the primary data source for the AI agent, which refines user queries, retrieves relevant information, and generates replies. The system prioritizes knowledge sources, weighting starred conversations highest, followed by historical tickets and general documents, ensuring contextually relevant and efficient responses.

Quick Start & Requirements

Deployment is facilitated via Docker. Key prerequisites include Bun ≥ 1.2.16, PostgreSQL with the pgvector extension, and MinIO (or S3-compatible object storage). AI capabilities necessitate OpenAI or FastGPT credentials. Local development involves cloning the repository, installing dependencies with bun install, configuring a .env.local file with database and storage URLs, and running database migrations via bun run migrate within the server directory. The primary Docker run command is: docker run -d --name tentix -p 3000:3000 --env-file ./.env.local tentix:latest.

Highlighted Details

• "10x Efficiency" claim for response time, human intervention, and customer satisfaction.
• AI workflow powered by LangGraph for analysis, query generation, retrieval, and answer generation.
• Extensible architecture supporting modular pluggable extensions and multi-channel notifications (Feishu supported).
• Unified Vector Knowledge Base built on PostgreSQL + pgvector, allowing configurable AI agent parameters.

Maintenance & Community

The repository outlines contribution guidelines following Conventional Commits standards and emphasizes code review via Pull Requests. Support is directed towards GitHub Issues. No specific community channels (e.g., Discord, Slack) or details on maintainers/sponsorships are provided in the README.

Licensing & Compatibility

The software license is not explicitly stated in the provided README, which represents a significant gap for due diligence regarding usage rights and compatibility.

Limitations & Caveats

The project is actively under development, with a roadmap detailing planned features such as comprehensive admin and agent dashboards, advanced analytics, prompt customization, and workflow orchestration, indicating these are not yet implemented. Specific infrastructure requirements (PostgreSQL with pgvector, MinIO) must be met. The absence of a stated license poses a potential adoption blocker.