Inspiration
One of our team members works a lot with stock brokers as a quant. He felt there is potential for AI to disrupt stock breakers, providing faster execution and lower commission.
What it does
It takes buyers’ and sellers’ bid/ask orders and negotiations for option trading in voice, and processes and matches/executes deals.
How we built it
We built it using Python and Java, with speech-to-text models, large language models, text-to-speech models, etc.
Challenges we ran into
The capabilities that AI interacts naturally with humans
Accomplishments that we're proud of
We successfully built an AI tool which can handle complex human interaction, and significantly improve market efficiency.
What we learned
Agile development, quick iteration.
What's next for AI voice agent for stock Broking
Include a operation function that check compliance (e.g. short selling restrictions)
More intelligent algo to search for potential counterparties
Market summary/update functionality
Agents that represent traders in price negotiation
Built With
- amazon-web-services
- elevenlabs
- fastapi
- html
- openai
- python
Log in or sign up for Devpost to join the conversation.