AI Chatbot & Voice Agent Developer Needed (RAG + n8n Automation)

We're looking for an experienced AI developer to build a conversational AI system for our business that combines chatbot/voice capabilities with automated workflows. What we need: - We want a chatbot (and potentially a voice agent down the line) that answers questions using our own documents and knowledge base — not generic AI responses, but accurate answers pulled from our actual content using retrieval-augmented generation (RAG). This needs to be connected to a vector database and set up so it stays accurate as we update our content over time. Alongside the chatbot, we need automation built using n8n (or a similar tool like Make/Zapier if you have a strong reason to recommend it) to connect the AI system to our existing tools — this includes things like updating records in our CRM, sending notifications, and triggering follow-up actions based on what happens in a conversation. If you have experience building voice AI agents (speech-to-text, LLM reasoning, text-to-speech, ideally with phone integration via Twilio), that's a big plus, since we may want to expand into a voice-based version of this system in a later phase. What we're looking for in a candidate: - Proven experience building RAG-based chatbots (not just calling an LLM API directly) - Hands-on experience with vector databases (Pinecone, Weaviate, Chroma, or similar) - Strong experience with n8n or comparable automation/orchestration tools - Experience with OpenAI, Anthropic, or similar LLM APIs - Bonus: experience with voice AI (Whisper, ElevenLabs, Twilio) or multi-agent frameworks like LangChain/LangGraph - Clear communicator who can explain technical decisions in plain language - Comfortable working async, with regular progress updates Process: Please include examples of similar RAG chatbots, voice agents, or n8n automation projects you've built (links, demos, or case studies), along with a short note on how you'd approach connecting the AI layer to our existing tools. Budget/Timeline: Open to discussing based on scope — happy to start with a smaller pilot (e.g., a single chatbot connected to one data source) before expanding into the full automation and voice capabilities.