Realistic AI Assistant w/ Voice, Memory & Internet

Project Brief: Personal AI Assistant — Voice, Memory & Live World Access Project Overview We are looking for an experienced AI/full-stack developer to build a personal AI assistant powered by the Anthropic Claude API. The goal is to create a fully interactive, voice-driven AI companion that feels like a real, intelligent presence — not a chatbot. It should: • Be accessible and conversational 24/7 • Respond word by word in real time like a natural human conversation • Remember everything about the user across all sessions • Have access to live world information at all times • Feel like talking to an actual intelligent being, not a tool Full Project Scope 1. Voice Conversation Interface • Microphone input so the user speaks naturally — no typing required • Real-time speech-to-text transcription using Whisper by OpenAI or Deepgram • Streaming AI responses via the Anthropic Claude API — word by word output, not delayed full responses • Natural human-sounding voice output using ElevenLabs or OpenAI TTS • Conversation must feel fluid and instant — like talking to a real person with no robotic pauses 2. Persistent Memory System • The AI must remember the user across every session, indefinitely • Build a memory database (e.g. PostgreSQL or Supabase) that stores: • User preferences, personality, and personal details • Full conversation history over time • Important context the user shares — life events, habits, goals • Memory is injected into every conversation so the AI always knows who it's talking to and builds on past interactions 3. Live World Access • The AI must have real-time access to the internet • Integrate web search so it can answer questions about current events, news, weather, markets, anything happening right now • The AI should feel present and aware of the world — not frozen at a knowledge cutoff date 4. Always-On Availability • The system must run 24/7 — reliably and without interruption • Hosted on a stable server so the user can access it any time of day or night • Fast response times are a priority — latency should feel minimal 5. Interface • Clean, simple app interface — desktop and/or mobile • Easy to use daily without friction • Basic settings panel to manage memory, voice preferences, and personality settings - Technical Requirements: • The ideal developer must have hands-on experience with: Anthropic Claude API — LLM integration and prompt engineering • Voice pipelines — speech-to-text and text-to-speech (Whisper, Deepgram, ElevenLabs, OpenAI TTS) • Backend development — Node.js or Python • Database design — persistent memory and conversation storage • Server deployment — hosting a 24/7 always-on application reliably - What We're Looking For • Someone who has built real AI-powered applications — not just experimented • Strong communicator who gives honest updates in plain English • Reliable and transparent about what is and isn't possible within budget | Ideally someone genuinely passionate about AI and conversational systems - How to Apply Please send: • A short introduction about yourself • 1–2 examples of similar projects you've built (voice AI, LLM-powered apps) • Your proposed approach to this project • Your hourly rate or fixed project quote • Your estimated timeline This project has long-term potential. A developer who delivers well will be first in line for future phases and ongoing work as the project grows.