SmartBank CRM AI Assistant

Data Visualization during transaction analysis in real time
Call summaries and tasks performed stored in database for reference
Flagged transactions and reasons being stored in the database

Inspiration

Our inspiration came from the growing need for efficient and seamless customer service in the banking industry. We envisioned a system where AI could handle routine tasks, provide instant information, and offer deep insights through data analytics, ultimately transforming the customer experience.

What it does

Our AI-Powered Banking CRM Solution allows customers to perform essential banking tasks such as updating information, blocking cards, and flagging fraudulent transactions through voice commands. It provides instant answers to general banking queries and loan options from a vector database, and offers data analytics and visualization to analyze transactions and assess credit card upgrade eligibility. We also store call summaries and tasks done within the database for future use.

How we built it

Employed OpenAI Whisper for real-time speech-to-text conversion.
Integrated a vector database for storing and retrieving customer information and bank-related data.
Developed and trained our own semantic router to accurately map and route tasks based on customer queries.
Implemented Pie chart data vizualization using ReactJs
Ensured a high level of security to protect sensitive customer information by routing unrelated queries away
Focused on iterative development and continuous improvement to refine the system's capabilities.

Challenges we ran into

We faced several challenges, including ensuring the accuracy and responsiveness of the voice recognition system, integrating the vector database seamlessly, and creating effective data visualization. Another challenge was maintaining a high level of security while allowing for quick and easy task execution.

Accomplishments that we're proud of

We are proud of creating a robust AI system that can handle a wide range of customer service tasks efficiently. Our successful integration of voice recognition, NLP, and data analytics into a single platform demonstrates the potential of AI in transforming customer service.

What we learned

Through this project, we learned the intricacies of combining various AI technologies to create a cohesive and functional solution. We gained insights into the importance of user experience in AI applications and the critical role of security in handling sensitive information. The project also taught us the value of iterative development and continuous improvement.

What's next for us

Right now this application can perform complex data processing and analytics in real time over a call. We want to modularize some of the features so that implementing similar solutions for other industries or business can also be done in an easy way. We really see this technology taking off in the Retail and Food Delivery sectors since the tasks being performed in customer service are well within the range of this application.