Skip to content

N0-man/ai-assistant

Repository files navigation

Contextual AI Assistant

The goal is to build cohesive or contextual smart agents leveraging LLM.

Prerequisite

  1. install dependencices
pip install -r requirements.txt
  1. create a .env file with OPENAI_API_KEY=<your key>

KAI (Knowledge AI Assistant)

KAI is mostly an LLM over set of documents (pdf, doc, html, epub etc). It leverages Embedding-based search to answer question from the context (documents).

TODO: Test other file formats for context e.g. ppt, doc, html, epub etc

Add some pdf's for context in ./docs folder
reference: OpenAI cookbook

Run

python kai.py clean --all


BAAI (Business Analysis AI Assistant)

BAAI can help breaking down business features into Epics > Stoires > Acceptance cirterias.

TODO: extend the solution to groom stories with more validations and edge scenerios

baai_context.br.py have top 5 features on building Singapore Rental car system extarcted from GPT. The questions that were asked to GPT were (1) what are the probelms with Singapore rental car system (2) Given above context what are the top 5 features to implement
reference: deeplearning.ai’s ChatGPT Prompt Engineering for Developers course

Run

python baai.py clean --all


Simple REST API for Q&A ChatBot based on PDF files (basic version)

This is just an alternate approach to use FastAPI to expose the application as REST endpoint

Run

python main.py clean --all

test API

Pulse Check

curl -X GET http://localhost:8001/pulse

Ask Question

curl -X POST -H "Content-Type: application/json" -d '{"source": "unknown", "question": "What is boo?"}' http://localhost:8001/ask

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages