Project Title: Build RAG Chatbot from Google Sheets Database – “Ask AI” Widget for Website/Landing Page Budget: $400–$500 Timeline: 7–10 days Project Overview I am looking for an experienced developer to build a Retrieval-Augmented Generation (RAG) chatbot and embed it as an "Ask AI" chat widget. The chatbot must answer user questions exclusively from my proprietary data source—a Google Sheets file—to prevent external web access and hallucinations. The Google Sheets file contains approximately 1,000–2,000 rows of scraped website content. Key data columns include: ID, type (article/video/podcast/etc.), title, date published, source name, source URL, word count, language, source topic, primary theme, secondary themes, short summary, long summary, and tags. Goal: Create a clean, fast, and functional chat widget that enables users to ask questions and receive answers based only on the content within the provided Google Sheet.Required Features Vector Database Setup: Convert the Google Sheets data into a searchable vector database. Preference for free/local options: Chroma, Pinecone, or Weaviate. AI Engine Integration: Utilize a fast, modern AI engine for the RAG pipeline. Options: Groq, LlamaIndex (preferred), or OpenAI. Chat Widget Development: Embed a clean, responsive chat widget on a simple landing page I will need for the project. Technology Recommendation: Streamlit, Tidio, or custom HTML/JS (developer's recommendation). Source Attribution: Implement a "Show sources" functionality. When an answer is provided, the user must be able to see the specific rows (titles, URLs, and/or summaries) from the Google Sheet that the answer came from. Nice-to-Have Features Mobile-friendly design for the widget and landing page. Basic privacy controls (e.g., no unnecessary third-party data sharing). Exportable chat logs (CSV or similar). Deliverables Fully working demo on a live web host (on the provided landing page). Complete code repository (GitHub private repo or zip file). Development of a simple landing page to host the widget. One 60-minute walkthrough call (record if possible) to hand over the project. Skills Needed Python RAG pipelines (Experience with LlamaIndex highly desired) Vector Databases Basic Frontend Development (HTML/JS/React) Please reply with your proposed solution, rate, estimated timeline, and links/screenshots of any past RAG chatbot projects.