CortexMind is a full-stack AI-powered chat application that simulates the functionality of modern large language model (LLM) platforms like ChatGPT and DeepSeek. This is a project that demonstrates several full-stack development skills, including real-time message streaming, authentication, API integration, and data persistence.
When a user clicks on a chat session, a request is sent to the backend to fetch all messages for that session, populating the page with the existing chat history.
Controls the temperature parameter of the DeepSeek-chat model, allowing users to adjust response creativity and style. More information can be found here: DeepSeek Parameter Settings
- Frontend: React, Tailwind CSS, Zustand
- Backend: Express.js, Node.js
- Database: PostgreSQL
- Authentication: Auth0
- AI Integration: DeepSeek LLM API
-
🔹 Real-Time AI Chat
Integrates with DeepSeek'sdeepseek-chatmodel to stream conversational responses incrementally. -
🔹 Responsive UI
Built with React and Tailwind CSS for a sleek and responsive chat interface. -
🔹 User Authentication
Secure login and session handling powered by Auth0, with chat history tied to user accounts. -
🔹 Persistent Storage
All messages and sessions are stored in PostgreSQL.
- Users log in through Auth0.
- Frontend sends prompts to the backend.
- Backend communicates with DeepSeek’s API, streams responses in real-time, and updates the database.
- Stored messages are retrieved from PostgreSQL per user session.
- Building real-time, stream-based chat UX
- Managing auth workflows across frontend and backend
- Structuring scalable database schemas for user-generated content
- Handling full-stack API integrations (LLM, Auth0, database)
This project was created for portfolio and learning purposes. It is not intended for production use and may lack certain security or scalability optimizations.