A solo developer built an AI chatbot that shows its sources

One developer built an AI chatbot designed for company documentation and help centers. Every answer the chatbot gives comes with a link to the exact document it pulled from. This makes it easy to check whether the answer is correct.

Most AI chatbots generate answers from scratch, which means they can confidently say things that are wrong. This chatbot works differently: it only answers based on documents you upload (like product manuals or FAQs), and it shows which document — and where in that document — each answer comes from.

This approach is called RAG (retrieval-augmented generation). By connecting your own internal docs or help center content, customers or staff can find accurate information faster without wading through pages manually. This is an indie side project, so it would need evaluation before use in a production environment.

Key points

  • Every answer includes a citation so you can verify the source
  • Works with your own documents such as manuals, FAQs, or wikis
  • Reduces AI 'hallucination' — where the AI invents facts that aren't real
  • Built by a solo developer; not yet a polished commercial product

Quick term guide

AI chatbot
A computer program that replies in a chat like a person.
AI chatbots
Software you can talk to that uses AI to answer or help with tasks.
RAG (Retrieval-Augmented Generation)
A technique where an AI searches an external knowledge base for relevant information before generating its answer
retrieval
The step where a system finds the most relevant text for a question.
side project
A small project someone builds outside their main job or main business.
valuation
The amount investors think a company is worth.
production
The live version of a service that real users use.
hallucination
When AI makes something up and presents it as a real answer.
Read original