Description
Summary Job Title: AI Conversational Agent Developer for Vocal Coaching (WhatsApp) Project Overview We are looking for an experienced developer to build a production-ready Conversational AI Agent for our vocal coaching business using the official WhatsApp Business Platform/API. We do not want a traditional rule-based or rigid button bot. We need an intelligent agent powered by LLMs that can comprehend context, reason, answer prospective student inquiries dynamically, qualify leads, and coordinate bookings. Crucially, the agent must rely on a flexible knowledge base that we can easily update and expand over time. Key Responsibilities • Develop Conversational AI Agent: Build an intelligent WhatsApp agent using the official WhatsApp Business Platform/API and advanced LLMs (OpenAI, Claude, or Gemini). • Dynamic Knowledge Base (RAG): Implement a retrieval architecture (like vector search or similar) that allows us to easily feed the agent with new information, FAQs, pricing, and updates without modifying code. • Operational Workflow (No CRM currently): Since we do not use an active CRM yet, the developer must design a clean alternative system (using Google Sheets or a similar flexible database) to capture lead data, store conversation context, and track booking statuses. • Automation Workflows: Automate responses regarding singing lesson formats, teacher profiles, methodologies, and scheduling. • Lead Qualification & Booking: Program the agent to ask the right qualifying questions and guide prospective students to book a trial class. • Human Handoff: Implement intelligent routing to flag and escalate conversations to a human representative when needed. Required Features • Updatable Knowledge Management: An easy way for us to add/edit the information the agent uses to answer questions. • Lead Capture & Storage: Automated logging of student interactions and lead details into Google Sheets or an accessible backend. • Calendar Integration: Seamless integration with scheduling tools (like Calendly) to handle appointments directly via WhatsApp. • Human Escalation Logic: Clear triggers to hand off complex cases to a human. • Modular Design: Built cleanly so that we can plug in a CRM later as our business scales. Preferred Tech Stack & Experience • Official WhatsApp Business Platform / Cloud API • Automation platforms: n8n or Make.com • LLMs: OpenAI GPT-4o, Claude, or Gemini • Experience building RAG (Retrieval-Augmented Generation) systems or dynamic knowledge bases. • REST APIs and Webhook integrations. What We’re Looking For We want a scalable, clean architecture and a developer who can provide clear documentation on how to maintain and update the agent's knowledge base, followed by a short handover session.