AI Developer for Offline Voice Agent

We are looking for an experienced AI developer to build an AI voice agent with a desktop application that runs efficiently on a local computer without relying on cloud APIs. Project Requirements The AI voice agent must: • Operate completely offline on a local machine. • Communicate through voice-to-voice interaction with natural, low-latency responses. • Use a predefined Excel knowledge base (vocabulary and data) as its primary source of information. • Strictly follow developer-defined instructions and business rules. • Never generate answers outside the provided knowledge base or instructions. If the requested information is unavailable, it should respond appropriately instead of hallucinating. • Allow users to upload updated Excel files so the knowledge base can be refreshed without modifying the application. Voice Features The application should include advanced voice capabilities, such as: • Accurate voice activity detection (VAD) • Noise suppression/cancellation • Automatic detection of user silence • Natural turn-taking during conversations • Fast speech recognition and speech synthesis • Low response latency for smooth conversations AI & System Requirements The solution should: • Use an efficient RAG pipeline. • Be optimized for CPU-based execution while maintaining good response speed and accuracy. • Be designed with hardware limitations in mind, including memory and processing constraints. • Have a modular architecture that makes future updates and maintenance straightforward. • Include a clean, user-friendly desktop interface. Required Skills The ideal developer should have proven experience with: • RAG pipeline development • LangChain • LangGraph • Agentic AI systems • Offline voice agents • Speech-to-Text (STT) and Text-to-Speech (TTS) • Local LLM deployment and optimization • Workflow and AI pipeline optimization • Desktop application development (Python, Electron, or similar) What We're Looking For We are looking for a developer who values quality over shortcuts. The ideal candidate should understand both AI engineering and system optimization, ensuring the application is reliable, responsive, and efficient on local hardware. Experience building offline AI assistants or voice agents is highly preferred.