title	PersonalChatbot
emoji	💬
colorFrom	yellow
colorTo	purple
sdk	gradio
app_file	app.py
pinned	false
hf_oauth	true
hf_oauth_scopes	inference-api

Based on Prof. Paffenroth's original CSDS553 Demo chatbot, added changes include:

File upload and context integration Response time measurement and display

Running with Docker

---

Step 1 — Build the image

./build.sh

Builds the Docker image and tags it as mlops-group8.

Step 2 — Start the containers

./run.sh

Starts the app and Grafana containers with the following port bindings:

Service	URL
Frontend (Gradio)	http://localhost:22111
Backend (FastAPI)	http://localhost:22112
Grafana	http://localhost:22113

Step 3 — Expose via devtunnel

Make sure you are logged in first:

~/bin/devtunnel user login

Then start the tunnel:

./tunnel.sh

devtunnel will print public URLs for each port in the terminal.