DocsLM
Upload up to 9 PDFs and get back a deployed website with an embedded chatbot that answers questions from your documents — live at a public URL in minutes.
Challenge
Teams spend hours reading through lengthy PDFs, reports, and documentation, and traditional file storage tools just hold the files — they don't help anyone understand them.
Approach
DocsLM runs a Django REST API for uploads and webhook orchestration, with n8n pipelines driving the multi-step workflow: upload, process, generate, deploy. Two LangGraph agents do the intelligence work — an Architect Agent runs an agentic RAG loop over the uploaded documents until it has extracted a complete site structure, and a Designer Agent generates the HTML/CSS site with an embedded chatbot widget. A Deployment Agent pushes the result to Vercel's API and returns a live public URL, giving each user their own site plus chatbot.
Outcome
Up to 9 PDFs become an interactive, shareable website with a working chatbot in minutes, with Django handling the UI layer and n8n handling AI orchestration as clean, separated concerns.