Payment (Read this first) This is a fixed bid, performance-based project. • Small upfront (minimal) • Majority paid only when first customer is live and system is working • Bonus per clinic that goes live If you are not confident you can deliver, don’t apply. What we are building: We are building a voice AI platform for dental clinics in the US. This will automate: • front desk calls • insurance verification • claim follow-up • scheduling • payment collection Cost requirement (very important): System must run at $0.02–$0.03 per minute voice cost (target) If your architecture is expensive, this will not work. You must know how to: • keep token usage low • design efficient prompts • handle streaming voice properly • keep latency and cost under control What you are actually building: This is not one bot. You are building a framework that supports multiple voice agents. System must support: • inbound + outbound calls • IVR navigation • IVR learning (automatic) • talking to live agents • structured data capture • reusable workflows AI agents framework and agents to build 1. AI Receptionist • answer calls • capture patient info • FAQs • appointment requests • SMS confirmations 2. Insurance Verification Agent • call payers • navigate IVR • verify eligibility • learn payer flows • talk to live reps • extract structured data 3. Claim Follow-Up Agent • call payers • navigate IVR • talk to reps • get claim status, denial, next steps • log structured output 4. Recall Agent • call patients for recall • confirm / reschedule • send reminders 5. Revenue Agent • call patients with balances • send payment links • capture responses • escalate when needed IVR learning (critical) System must: • detect IVR menus • learn call paths • reuse best path next time • improve over time No hardcoded flows. MVP deliverables • AI receptionist working • IVR navigation working • insurance + claim follow-up prototype • transcripts + logs • admin dashboard • deployed test environment Skills required • Voice AI (Twilio, Vapi, etc.) • OpenAI / Claude • Python or Node backend • APIs and automation • real-time systems To apply Send: 1. GitHub 2. 2 real systems you built 3. voice AI demo (if you have it) 4. how you would build: • IVR navigation • IVR learning • claim follow-up • low cost system 5. timeline 6. screen shot that shows per min call cost If you don’t have real examples, don’t apply. Screening question How would you build an AI system that can: • navigate IVR • talk to live agents • extract structured claim status • keep cost very low (can or templated answers will be rejected immediately) Important Start your proposal with the word: IVR If not, I won’t read it. Who should not apply • chatbot-only developers • no voice experience • slow enterprise consultants • people who cannot ship fast Final note: We are building a real product, not a demo. If you deliver, there is long-term work.