I built a RAG-based AI chatbot SaaS for document Q&A — DirayahAI

Hey IH,

I've been building DirayahAI over the past few months alongside

my day job. It's a multi-tenant SaaS platform that lets businesses

upload their documents and get an embeddable AI chatbot that answers

questions strictly from their content.

The problem I kept seeing: companies have HR policies, compliance

manuals, SOPs — all in PDFs nobody reads. Employees and customers

ask the same questions repeatedly. The answers exist, they're just

inaccessible.

Tech stack:

- .NET 8 backend with RAG pipeline

- PostgreSQL + pgvector for embeddings

- OpenAI / Claude as LLM providers

- React + Vite embeddable widget

- Multi-tenant with per-tenant document isolation

Current status:

- Platform is live at dirayahai.com

- Free trial available, no credit card required

- Working on first paying customers

Happy to answer questions about the build, the stack, or the

go-to-market approach.

https://dirayahai.com