Smart Proactive AI Agent for Customer Engagement

Using Generative AI

Welcome to my experimental platform, a personal project dedicated to enhancing customer engagement through proactive AI. This innovative solution leverages Generative AI and Large Language Models (LLMs) to initiate and manage conversations via SMS and voice. By scheduling outbound communications and interacting intelligently with responses, it aims to set new standards in proactive customer interaction. This website serves as my sandbox for exploration and experimentation, showcasing how AI can transform proactive customer engagement with minimal training and setup.

Please note, this platform is not affiliated with any company and is intended solely for personal use.

For a demo or access to the system, please contact me.

Use Cases Explored

Hi Mary, This is Dr. Philip from Phi Dental Clinic.You have a dental appointment scheduled on Thursday May 9 at 3 PM.You can confirm, reschedule, or cancel your appointment here.

Dental Appointment

A dental clinic utilizes reminders to encourage patients to confirm, reschedule, or cancel their appointments, effectively reducing no-shows.

Hi Angie, Your washing machine installation is scheduled for Friday, May 10 at 10 AM. You can confirm, reschedule, or cancel your appointment here.

HOME APPLIANCE INSTALLATION

Engage with customers to ensure a seamless setup experience, confirming their availability and providing rescheduling options.

Hi John, It’s onDimi Bank. Your loan quote is has been pre-approved. Would you like to go over the details with a Lending Specialist?

Loan Application Process

Proactively reach out to loan applicants about their online pre-approval, address any questions, and schedule a meeting with a lending specialist.

ADMIN AND DASHBOARD

Comprehensive System Management

The Admin panel enables you to configure and manage all aspects of the system. Watch the Video

Use Cases: Create new use cases or choose from existing templates.
Plans and Actions: Outline the sequence and frequency of actions within each plan.
Conversations: Customize the conversation flow of your plans to better suit individual needs.
Prompt Engineering: Craft and refine prompts to guide AI interactions.
Intents and Guardrails: Establish rules and parameters for interactions, choosing from predefined responses, generative responses, or knowledge-based outputs to ensure clear and effective communication.
Knowledge Base: Populate the system with specific business-related information to enhance the relevancy and accuracy of responses.

DYNAMIC CADENCE

Dynamic Plans

A way to not only schedule but also refining the cadence of conversations based on what the customers really need.

Schedule Conversations: Planning ahead, setting up future conversations with customers to ensure we keep the conversation going.
Modify Plans: Plans are redefined on the fly based on the customer responses and the business needs.
Stop or Clear Planned Interactions: When the needs changed, quicly adjust or cancel the planned interactions.

YOUR RESPONSE - YOUR CHOICE

Predefined, Generative, or Knowledge Base

Once the true needs or inquiries of a customer are identified, the GenAI concept allows for responses in one of three tailored ways:

Predefined Responses: These responses are predefined, and preapproved ensuring accuracte response.
Generated Responses: When GenAI excels in specific business area, it will take the lead to create dynamic and contextually appropriate responses.
Knowledge Base Responses: For business related inquiries, Retrieval-Augmented Generation (RAG) technolgoy is used.

GUARDRAILS

Ensuring Focused and Safe Conversations

Implementing strict guardrails ensuring all interactions remain business related and safe preventing any deviation from business intended goals. Here’s why they’re important

Retrieval-Augmented Generation (RAG)

Boosting Accuracy in Business-Specific Responses

Implementing Retrieval-Augmented Generation (RAG) ensures LLMs provide accurate and up-to-date business-specific responses by leveraging external knowledge bases:

Knowledge Base Representation: Business data is converted into numerical representations (embeddings) using AI techniques. These embeddings are stored in a vector database for efficient retrieval.
User Query Processing: The user's query is converted into a vector representation and matched with the vector databases to retrieve relevant data in context (based on similarity/relevancy/relatedness).
Response Generation: The retrieved data and the user's query are fed into the LLM. It uses prompt engineering to allow the LLM to generate an accurate response.

INBOUND & OUTBOUND CALL

Enhancing Voice Call Handling

Experiementing the use of GenAI, Whisper model for speech recognition, Google TTS, and openAI LLM to enhance the conversational experience in voice calls. Watch the Video

Personalization: Personalizes interactions by recognizing callers
Call Topic: Detecting the call topic to be discussed in a voice call.
Answering queries: Easily and accurately answer queries over the phone without the need fot an agent involvement.
Transcription: Transcribing the call on the fly while the call is going on.
Scheduling call/appointment: Scheduling, Rescheduling or canceling a call or appointment. Transfering to agent if required.
Setting follow up actions: Setting reminders or a new plan based on the outcome of the call.