Laurel AI Chat Bot

Laurel AI Chat Bot is a full-stack AI chat application focused on reliable session management, model selection, request tracing, and reproducible local development.

Repository: github.com/muzaffercanan/laurel-ai-chat

AI chat products need more than a prompt box. A useful system should keep conversation history, expose model choices, trace requests, and remain easy to run locally.

I developed the frontend, backend API layer, persistence flow, observability integration, and Docker-based local setup.

• React and TypeScript
• Express.js
• SQLite
• OpenTelemetry
• Jaeger
• Docker and Docker Compose

• Built RESTful APIs for session management and chat history persistence.
• Added dynamic model selection rather than tying the UI to a single model.
• Integrated OpenTelemetry and Jaeger for end-to-end request tracing.
• Containerized the application for reproducible local development.

• Keeping frontend state aligned with persisted chat sessions.
• Designing a backend API that could support multiple model choices.
• Making observability useful without adding unnecessary complexity.

The project demonstrates full-stack AI application architecture with practical backend concerns: sessions, persistence, model routing, tracing, and local deployment.

AI application quality depends heavily on the surrounding product infrastructure. Session design, observability, and reproducibility matter as much as the generated response.