About

Description: We are building an AI-powered video survey tool. The system will allow the AI agent to run a live video call, guide users through a structured survey, capture photos, and generate a preliminary quote in real time. We are looking for an experienced full-stack developer (or small team) to build a browser-based MVP with the following features: MVP Scope Frontend: Web app (React preferred) WebRTC video call integration (Twilio/Daily/Agora) UI for AI-led survey prompts, progress display, and photo uploads Backend: Python/Node.js API for survey logic and quote generation Session and consent data storage (PostgreSQL or similar) Integration with AI services (LLM for dialogue management, optional Whisper/STT) AI Integration: Real-time speech-to-text (Whisper / Google Speech-to-Text) LLM prompts for guided survey and structured slot-filling Quote Engine: Rule-based engine to calculate estimates based on survey responses PDF quote generation with confidence range Optional (phase 2 / bonus): Computer vision for analyzing photos Key Requirements Proven experience building WebRTC video apps Experience integrating LLMs and/or AI services Strong frontend + backend skills (React + Python/Node.js) Familiarity with cloud deployment (AWS/GCP/Azure) and API integrations Ability to deliver working MVP within 2–3 months Clear communication and documentation. Aim will be to transfer MVP into an App should it gain traction.