AI Opportunity Center Chatbot

Team Members

Bradley Charles
Rae Maffei
Livan Hagi Osman
Mark Yosinao

Project Overview

We are designing a prototype for a secure, in-house chatbot system with the focus on compliance, security, data privacy, and protecting intellectual property that are typically at risk with AI chatbots. Our proposed solution will include research, a detailed project report, and a final deliverable of a simple, working prototype. If possible, we wanted it to have multilingual capabilities.

Requirements:

• Compliance with HIPAA, FERPA, etc.
• Protection for Intellectual Property
• Run on internal, in-house infrastructure
• Provide articulate and helpful responses based on approved knowledge
• Use authentication tools to secure interactions
• Maintain audit logs as per compliance regulations

Tech Stack

1. Frontend
   • TypeScript
   • Next.js
2. Backend
   • Python
   • FastAPI
3. LLM
   • Hugging Face (TinyLlama-1.1B-Chat)
4. Authentication
   • Basic token support (WIP)

Setup & Installation

Prerequisites

• Python 3.11
• Node.js 18+ (with npm)
• Git

1) Clone the repo

git clone https://github.com/OC-Chatbot/Secure-Internal-Chatbot-Design.git
cd Secure-Internal-Chatbot-Design

2) Linux install & run (bash)

Backend setup:

python3.11 -m venv .venv
source .venv/bin/activate
pip install -r backend/requirements.txt

Frontend setup:

npm install

Run both services (recommended single command):

./.venv/bin/python start_services.py

This starts uvicorn backend.main:app on port 8000 and next dev on port 3000. It also sets NEXT_PUBLIC_API_URL to http://localhost:8000/api if unset.

Manual alternative:

source .venv/bin/activate
export NEXT_PUBLIC_API_URL=http://localhost:8000/api
uvicorn backend.main:app --reload --host 0.0.0.0 --port 8000

# New terminal
npm run dev -- --port 3000

3) Windows install & run (PowerShell)

Backend setup:

python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r backend\requirements.txt

Frontend setup:

npm install

Run both services (recommended single command):

.\.venv\Scripts\python.exe .\start_services.py

Manual alternative:

.\.venv\Scripts\Activate.ps1
$env:NEXT_PUBLIC_API_URL="http://localhost:8000/api"
uvicorn backend.main:app --reload --host 0.0.0.0 --port 8000

# New PowerShell window
npm run dev -- --port 3000

4) Access the app

• Frontend: http://localhost:3000 (chat UI at /chat)
• Backend docs: http://localhost:8000/docs

Notes

• The backend loads the TinyLlama model from Hugging Face; first run requires network access or a pre-cached model in your HF cache.
• Configure NEXT_PUBLIC_API_URL if your backend runs on a different host/port. You can export it before npm run dev or set it in .env.local.