RAG using AWS Aurora, Langchain and Llama 3

Goal

Point to the location of the datasource - codebase, pdf, etc. and ask questions about it.

DEMO

---

Index

1. Sign-in Page
2. Landing Page
3. Chat Bot

Note - This project was built during a hackahton. Therefore, it is not production ready. You can use it for reference and build your own production ready solution.

---

Sign-in Page

• Mandatory user sign-In for user tracking

---

Landing Page

• Repository Details - holds access to all information about your current data source and RAG status.

  • Repo Name and branch - tells the name and branch of repository that is being given as context to the LLM.
  • Actions -
    • Open Chat Bot - Link to open chat-bot.
    • Re-Sync - When data is changed at the source, a user can re-sync and provide fresh context to LLM.
    • Delete - Delete the existing chat-bot to free memory.

• Open Form Button -

  • Use it to give location to the new datasource for chat-bot to train on.

---

Chat Bot

Directly chat with your datasource -

---