Building a Multi-Agent System for Real-Time Conversations Between Humans and Als

The diagram of architecture of the system design

Al and LLMs have shown significant advancements in natural language understanding and generation. However, there is still a gap in the market for a comprehensive solution that can effectively synthesize information from multiple sources and present it in an intuitive, user-friendly manner.

The project aims to create a unified product that enables agent-to-agent interaction. This ‘Al reporter’ will be capable of analyzing articles stored in a database, extracting relevant information, and presenting it to the end user in a clear, concise, and visually appealing manner based on their specific queries. To enhance the user experience (UX) of the conversation, the system will incorporate speech generation and captions, making the interaction more engaging and accessible.

Hin Chi Kwok
Hin Chi Kwok
Student in IEOR and Computing

I strongly believe that the 3Cs - CURIOSITY, CHALLENGE, and CHANGE - will shape my future, help me achieve my goals, and positively impact my community. I am passionate for technology, and I am here to share my journey of rediscovering my passion for STEM. Despite facing gender stereotypes and societal expectations from selecting IT as my career before, I pursue my CURIOSITY and participate in research projects and competitions. I embrace CHALLENGES, seek innovative solutions, and am actively involved in academic exchanges and entrepreneurship. My ultimate ambition is to translate my research into real products, contribute to make CHANGES in the science and technology industry, and inspire others to pursue a career in STEM.