Voice AI Developer — Full-Scope Phone Agent for Veterinary Clinics (Detailed Spec)

We build automation systems for veterinary clinics, a sector currently overwhelmed by phone volume and short-staffed teams. We need a developer who can build and deliver the complete system end-to-end — from basic call handling through full calendar automation and premium features. Below is a detailed breakdown of what the agent needs to do at each stage. 1. Incoming call handling • Detect when a call should be routed to the AI agent (line busy, after hours, high call volume, or staff unavailable) • Greet the caller and identify the clinic by name • Answer common questions accurately: opening hours, pricing ranges, clinic address, parking/access info, accepted payment methods • Detect if the caller is describing an emergency (specific symptoms/keywords) and immediately route the call to a live person or emergency line, skipping all other steps • Handle callers speaking French, German, or English, and respond in the same language automatically 2. Appointment request intake (non-emergency) • Ask for and record: caller name, pet name, species/breed if relevant, reason for visit, phone number, preferred day/time • Read the clinic’s calendar (read-only access) to check real availability during the call • Offer 2–3 actual open time slots that match the caller’s preference • Let the caller pick a slot, or ask to be called back if none work • Confirm the details back to the caller before ending the call 3. Booking confirmation • Automatically write the confirmed appointment into the clinic’s calendar (read + write access), with all relevant details attached (pet name, reason, contact info) • Send a confirmation message to the caller (SMS or email) with the date, time, and any pre-visit instructions • If no slot could be booked automatically, forward the request to staff (SMS, email, or shared dashboard) for manual follow-up 4. Pre-appointment reminders • Automatically message or call clients a set number of hours/days before their appointment (configurable per clinic) • Allow the client to confirm or cancel directly through that reminder (SMS reply or short call) • If a client cancels, automatically check the waitlist (if one exists) and offer the freed-up slot to the next person in line • If a client doesn’t respond to a reminder, flag it for staff follow-up 5. Reporting • Track and log: total calls handled, calls resolved without human involvement, appointments booked, appointments confirmed via reminder, no-shows prevented, average call handling time • Generate a simple monthly summary (dashboard or emailed report) per clinic 6. System requirements • Built on proven platforms (Vapi, Retell AI, or Bland AI) for the voice layer, Twilio for call handling, and n8n or Make for workflow orchestration — not a custom AI model built from scratch • Calendar integration via Google Calendar, Outlook, or a comparable API depending on what each clinic already uses • Configuration (clinic name, hours, tone of voice, phone number, language) should be adjustable through a simple setup process, not a code rewrite, so the same system can be redeployed for other clinics quickly • Clear error handling: if any step fails (API down, unclear caller input, etc.), the call should still end gracefully and route to a human when needed Ideal profile: • Demonstrable experience building complete, production-grade voice AI systems (portfolio required, not just prototypes) • Strong command of Vapi, Retell AI, Bland AI, and Twilio • Experience with calendar/booking API integrations (read and write) • Experience with multilingual voice flows • Comfortable building and documenting workflows in n8n or Make • Clear communicator, able to explain technical tradeoffs to a non-technical client, reliable with deadlines In your proposal, please tell us: 1. Your proposed architecture and tools for each part of the system above 2. Your price for delivering the complete system as described 3. Whether you’d recommend phasing the build (and how you’d phase it) or building it all at once, with reasoning 4. Any recurring monthly costs (hosting, per-minute call costs, third-party subscriptions) 5. A realistic delivery timeline 6. How you’d approach making this reusable/deployable for other clinics with minimal rework 7. Examples of similar full-scope projects you’ve completed, ideally including calendar automation and multilingual support