Hey Makers
We are excited to announce the winner of our sixth community challenge, focused on building AI-powered CX Solutions with Make and Voiceflow, @KaiHuxmann!
This submission stood out for its robust, clean architecture and for fixing a massive pain point in Customer Experience.
A massive congratulations to Kai, who built a rock-solid, multi-agent solution that transforms the “sucking” online chat experience into a secure, automated, and action-oriented front-door!
The Story: Fixing the Broken Customer Front-Door
Kai’s story addresses a pain point every company faces: Most online Chat Agents totally suck as they have no access to the right data. A lot of companies struggle with the very first step of talking to a customer. People reach out needing an invoice, help with a project, or looking for a service, but “most every experience is horrible for the user”.
Kai wanted to fix the core issue: creating a smart, automated way to greet users, understand what they need, qualify real leads, or securely verify existing customers. Traditional chatbots can answer basic FAQs, but they can’t actually do anything meaningful.
The Solution: A Secure, Multi-Agent Digital Front Desk
Kai’s solution connects three core elements to create a front-door experience that connects helpful conversation with real, automated actions in the background:
- Conversational Brain: VoiceFlow acts as the frontend and conversational brain for user communication.
- Automated Workflows: Make provides the solid automated workflows, clear separation of concerns, and data security.
- Secure Verification: It uses a multi-agent setup with secure OTP verification powered by Make.com, ensuring frontend agents don’t contain any customer data before the verification is complete.
The biggest challenge was getting the three-agent architecture to work seamlessly, ensuring each agent had focused prompts while maintaining the same tonality and name for a unified user experience.
What Our Champions Had to Say
Kai’s submission impressed our judge with its focus on security and maintainability. Here’s a summary of their feedback:
“Winner for me. Solves a clear problem that is often overlooked when implementing AI chats. The OTP verification is simple but very effective and cool to see.”
“I like the use of OTP and getting customer details securely for further context in the conversation. Generic AI chatbots are helpful but even more so with added context. His vision seems clear and with some polishing could be an awesome solution.”
“I wish every company that insists on having AI chatbots implements a variation of this flow.”
What’s Next for Kai?
The architecture is flexible by design, making it easy to swap logic or add new actions. Users who are authenticated can already self-service common requests like resending an old invoice, getting subscription info, or changing a billing address.
Kai’s next step is to introduce Calendar Booking for users who are qualified as a “Lead” against the ICP definition, ensuring teams mostly talk to pre-qualified leads.
This is what it’s all about—using technology to give customers a better experience without hiring more people. Congratulations again, Kai, on an incredible and inspiring submission! 