-
-
Chat page
-
Sign Up Page
-
Log In page
-
Forgot password page
-
Query and Resposne
Inspiration
The goal was to build a platform that allows users to chat with large language models (LLMs) while accessing real-time information, which could be useful for business decisions, research, or everyday tasks.
What it does
CurrentAI enables users to interact with a conversational AI in real-time. Users can ask questions, get responses, and have their conversation history saved securely. It also provides features like password management, user authentication, and the ability to export chat history.
How we built it
Front-end: Developed with Streamlit for a user-friendly interface that handles chat input/output and navigation between sign-up, login, and chat screens. Backend/Logic: Used LangChain to create the LLM-powered agent that processes user queries and generates responses. Database: Integrated with TiDB Serverless to store user information, queries, and chat histories, ensuring reliable and scalable data management. Session Management: Leveraged Streamlit session state to handle user sessions, login/logout flows, and storing real-time chat data.
Challenges we ran into
User Interaction and Persistence: Managing user data, chat history, and ensuring smooth interaction between users and the AI posed challenges. We addressed this by integrating TiDB Serverless, which helped efficiently store and retrieve user queries, responses, and session data, while keeping performance high.
Accomplishments that we're proud of
Scalable Solution: We're proud to have successfully combined Streamlit for the interface, LangChain for AI logic, and TiDB Serverless for database storage into a scalable and reliable system. Smooth User Experience: Despite the complexities of real-time data and multiple backend services, the system provides a seamless chat experience with minimal friction.
What we learned
How to coordinate various technologies like Streamlit, LangChain, and TiDB Serverless to create a cohesive application. Real-time AI Integration: The intricacies of integrating real-time data with LLMs to give users up-to-date responses.
What's next for CurrentAI
Enhanced Authentication: Improving the security and user login process, including social logins. Chat Sharing: Allowing users to share their conversations via email or other methods. Advanced Features: Adding more interactive and customizable features, such as voice inputs, multi-user collaboration in chat, and analytics.
Log in or sign up for Devpost to join the conversation.