Inspiration
The idea for "Cover With Care" was inspired by the number of unsatisfied customers calls we experienced first hand, where we felt unheard, misunderstood, or disrespected during customer support interactions.
We realized that while technical skills are often emphasized in customer service training, the soft skills—like empathy—are frequently overlooked. We aimed to bridge this gap, not just to improve customer satisfaction but also to empower customer service agents to handle challenging situations with emotional intelligence.
What it does
"Cover With Care" is an interactive, AI-powered training platform designed to improve the emotional intelligence of customer service agents. It simulates real-life customer interaction scenarios where the agent must respond. Our empathy scoring algorithm provides immediate feedback, allowing agents to improve their empathy. By focusing on empathy and active listening, our platform aims to raise the standards for customer care.
How we built it
We utilized Next.js as our full stack framework. To simulate customer intractions, we use a large language model chat agent, acting as a customer. We use another large language model agent as the empathy coach, which scores the responses of the agent, provides actionable feedback and suggestions for improvement.
Furthermore, to implement voice, we used APIs from AssemblyAI (streaming speech to text) and ElevenLabs (streaming text to speech).
We consulted experts available on site from Zurich Insurance to design the user interface and the scenario included in the demo.
Challenges we ran into
One of the main challenges was to accurately define "empathy" in a quantifiable manner and to tune the large language model responses. Generating a realistic customer interaction scenario was also tricky, as the large language model would often refuse to keep acting.
Time was another constraint, given that HackZurich had a limited timeframe of 40 hours for project completion.
Accomplishments that we're proud of
We're proud to have built a working prototype within the hackathon timeframe. Our initial testing with hackathon participants convinced us that the idea has potential to be pursued further.
What we learned
We learned about the immense importance of soft skills, especially empathy, in customer service settings, as well as the complexities involved in quantifying emotional intelligence.
What's next for Cover With Care
We plan to refine our algorithms based on the feedback and data gathered from pilot tests. We aim to add more industry-specific scenarios to make the training more comprehensive.
Ultimately, we will explore building this idea into a SaaS (Software as a Service) solution for businesses looking to invest in the emotional intelligence training of their customer support teams.
By prioritizing empathy, "Cover With Care" aspires to transform customer service interactions into positive experiences for both customers and support agents alike.
Built With
- assemblyai
- elevenlabs
- github
- javascript
- loom
- nextjs
- react
- typescript
- vercel
Log in or sign up for Devpost to join the conversation.