About

We are looking for an experienced backend engineer or small development team to build a voice-based AI calling system. This system allows callers to: • Dial a phone number • Connect in real time with an AI conversational voice agent • Interact using natural speech • Use keypad (DTMF) for simple menu navigation • Add or purchase usage minutes • Manage billing through secure card capture • Track usage time and costs • Interact with multiple “companion-style” personalities The system must integrate: • Twilio Programmable Voice (inbound calling, streaming audio, DTMF menus) • Realtime AI voice API (OpenAI Realtime or equivalent) • Stripe (subscriptions, one-time charges, saved payment methods) • Backend database (user profiles, usage tracking, billing state) This is a server-side engineering project only — no mobile app, no complex web app, and no UI/UX design. All details will be provided AFTER NDA is signed. ⸻ High-Level Responsibilities (Safe to share) • Build inbound call flow using Twilio Voice • Stream audio between Twilio and an AI voice model (low latency required) • Implement a DTMF menu system for simple options • Implement metered usage tracking (minutes consumed per call) • Integrate Stripe for payments (subscription + one-time purchases) • Store very minimal user profiles • Maintain clean, documented backend code • Provide deployment instructions Full specifications will be shared after NDA signing. ⸻ Required Skills (safe, non-specific) • Twilio Programmable Voice (must have actual experience) • Audio streaming pipelines (Twilio — Realtime AI) • Stripe integration (subscriptions + payment methods + webhooks) • Backend development (Node.js or Python preferred) • Secure handling of DTMF card entry • Experience with conversational AI or TTS/ASR systems is a big plus ⸻ Project Timeline Approx. 6–10 weeks, depending on scope finalization after NDA. ⸻ ❗IMPORTANT We will NOT share product details or documentation until: 1. You submit a proposal 2. You pass the screening questions 3. You sign an NDA