What's included

• Custom AI Agent (RAG or Voice)

  Description: A production-ready AI agent built to your use case—either a RAG agent for knowledge retrieval or a voice agent for real-time conversations. Includes: - Use-case specific agent logic - Prompt design and reasoning flow - Tool/function calling setup - Safety limits and guardrails Format: - Deployed agent (API, webhook, or app endpoint) - Source configuration files Revisions: Up to 2 revisions

• Knowledge Base & RAG Pipeline

  Description: A complete Retrieval-Augmented Generation pipeline that enables the agent to answer accurately using your data. Includes: - Document ingestion (PDFs, URLs, databases) - Chunking, embeddings, and indexing - Vector database setup - Retrieval optimization Format: - Live vector database - Sample test queries and outputs Revisions: Up to 2 revisions on retrieval logic

• Voice Agent Setup (if applicable)

  Description: A conversational AI voice agent capable of real-time speech-to-text, reasoning, and text-to-speech. Includes: - Call flow design - STT / TTS integration - Intent handling and fallback logic - Rate limiting and logging Format: - Live voice agent endpoint - Test call setup Revisions: Up to 2 revisions on conversation flow

• Automation & Orchestration (n8n)

  Description: n8n workflows to orchestrate the AI agent with external systems. Includes: - API integrations (CRM, ticketing, databases) - Logging and error handling - Triggered or event-based execution Format: - n8n workflows (.json export + live setup)

• Documentation & Handoff

  Description: Clear documentation to ensure your team can operate, extend, or maintain the system. Includes: - Architecture overview - How-to guide for updates - Prompt and data management instructions Format: PDF or Google Doc