shravv125-cmyk/Ask_question — reverse-engineered prompt

Build me a simple web app where someone can upload a PDF and then ask questions about it. The app should read the PDF, break the text into useful pieces, search for the most relevant parts when a question is asked, and then use an AI model to answer based only on the uploaded document.

Use Python with Flask for the backend and make a clean basic frontend with HTML, CSS, and JavaScript. It should feel like a small PDF AI assistant, with an upload area, a question box, and a place to show the answer. Use embeddings and vector search so the answers come from the document content, not random guessing. Use Groq with Llama for generating the final response, and make the API key configurable through an environment variable.

Please keep it beginner friendly and easy to run locally. Add clear setup instructions, handle errors like missing PDF or empty questions, and make the UI look neat enough for a first portfolio project.