Inspiration
Banks invest heavily in acquiring new customers, but the journey from identifying a potential lead to completing onboarding is often fragmented. Leads may come from campaigns, referrals, websites, or existing customer data, yet follow-ups are frequently handled manually across separate calling, messaging, CRM, and onboarding systems.
We were inspired to solve this gap by building a single intelligent platform that could manage the customer journey from start to finish. Instead of treating AI as only a chatbot or calling tool, we wanted to create a coordinated system that could identify prospects, engage them through their preferred channel, understand their needs, and guide them toward becoming fully onboarded banking customers.
What It Does Finora is an AI-powered customer acquisition and onboarding platform designed for banks.
The platform helps banks:
Identify and organize potential customer leads.
Prioritize prospects based on their profile, needs, and likelihood of conversion.
Contact customers through an AI calling agent powered by CALL-E.
Send personalized follow-up messages through a messaging agent.
Answer common questions about banking products and services.
Qualify leads based on their requirements and eligibility.
Collect necessary customer information and documents.
Track progress throughout the onboarding process.
Automate follow-ups and reduce customer drop-offs.
Customize communication, workflows, and branding for each bank.
Finora connects these steps into one continuous workflow:
Lead Identification → AI Engagement → Qualification → Follow-up → Onboarding Lead Identification→AI Engagement→Qualification→Follow-up→Onboarding The goal is not to replace human banking teams completely, but to automate repetitive interactions so bank employees can focus on complex cases, relationship-building, and final approvals.
What it does
We designed Finora as a modular, bank-configurable platform with multiple specialized AI agents working together.
The core workflow begins with a lead entering the system. The platform organizes and evaluates the lead, after which the voice agent can initiate a conversation using CALL-E. Based on the result of the call, Finora can trigger the messaging agent to send relevant information, reminders, or next steps.
The system maintains the lead’s status throughout the journey. For example, a lead may move through stages such as:
New lead.
Contact attempted.
Interested.
Qualified.
Documents pending.
Verification in progress.
Onboarding completed.
We built the project using a combination of:
A web dashboard for bank employees.
Backend APIs for managing leads, customers, workflows, and agent actions.
CALL-E for AI-powered outbound phone calls.
A messaging agent for automated customer communication.
AI-based workflow logic for deciding the next appropriate action.
A database for storing lead details, interaction history, and onboarding progress.
Configurable templates and settings to adapt the platform to different banks and their brand identities.
The architecture was designed to be flexible because each bank may have different products, eligibility rules, compliance requirements, communication preferences, and onboarding processes.
How we built it
One of our biggest challenges was coordinating multiple AI agents within a single customer journey. A calling agent and a messaging agent should not operate independently; each action needs to be based on the customer’s current status and previous interactions. Designing the workflow so that every agent had the right context required careful planning.
We also had to think about reliability. Phone calls may not be answered, customers may provide incomplete information, and messages may not receive a response. We therefore needed to account for retries, follow-ups, changing lead statuses, and cases where the workflow should be handed over to a human employee.
Another challenge was designing the product for multiple banks rather than a single organization. A solution that works for one bank may not work for another because banks differ in branding, products, customer segments, approval processes, and communication policies. We addressed this by making the workflows, prompts, messaging templates, and customer journeys configurable.
Working with financial workflows also made us more aware of the importance of privacy, security, consent, and responsible AI. Banking automation must be transparent and should avoid making unsupported promises or decisions. These considerations influenced how we structured the system and where human review remains necessary.
Challenges we ran into
We are proud to have built a complete customer journey instead of an isolated AI feature. Finora demonstrates how voice AI, messaging automation, lead management, and onboarding workflows can work together as one platform.
Our main accomplishments include:
Creating an end-to-end banking acquisition and onboarding workflow.
Integrating CALL-E to enable AI-powered customer calls.
Building a messaging agent for automated customer follow-ups.
Designing a configurable system that can be adapted to different banks.
Connecting customer interactions with lead and onboarding status.
Demonstrating how agentic AI can automate repetitive operational tasks.
Accomplishments that we're proud of
What we learned
This project taught us that building an effective AI product involves much more than connecting an AI model to an interface. The most important part is designing reliable workflows around the model and ensuring that each agent has the correct context, instructions, and limitations.
We learned how to:
Design multi-agent workflows for a practical business use case.
Integrate voice and messaging capabilities into one application.
Manage state across a long-running customer journey.
Build configurable products instead of one-off demonstrations.
Think about human handoffs, failure cases, and system reliability.
Design AI experiences for regulated industries such as banking.
Balance automation with privacy, security, transparency, and customer trust.
What's next for Finora
In the future, we plan to extend Finora with:
Integration with banking CRMs and core banking systems.
KYC and document-verification APIs.
Multilingual voice and messaging support.
WhatsApp and email communication channels.
Advanced analytics for conversion rates and onboarding drop-offs.
Human-agent handoff for complex conversations.
Consent management, audit logs, and stronger access controls.
Personalized product recommendations based on customer needs.
Continuous improvement using campaign and conversation analytics.
Our long-term vision is for Finora to become an intelligent, configurable growth layer for banks—helping them move from the first customer interaction to successful onboarding through one coordinated and trustworthy platform.
Log in or sign up for Devpost to join the conversation.