SECTION 1 — Why Deployment Matters

Right now, your AI app only runs on your computer.

That means:

• nobody else can use it
• nobody can test it
• nobody can share it

Real AI applications need deployment.

Deployment means:
making your application accessible online through a public URL.

This is one of the biggest transitions in becoming an AI developer.

You are moving from:
“local experiments”

to

“real usable applications.”

SECTION 2 — What Learners Will Build

In this lesson, we will:

• connect our project to GitHub
• upload our code
• deploy the app using Streamlit Community Cloud
• generate a public app link

The final result will be a working AI application anyone can access from a browser.

SECTION 3 — Project Setup

Folder Structure

deployable_ai_app

app.py

requirements.txt

.streamlit

secrets.toml

requirements.txt

streamlit

google-generativeai

app.py

Use the previously created working Gemini chat app.

Important Setup Note

Before recording:

• test app locally
• confirm Gemini API works
• confirm GitHub account works
• confirm Streamlit Cloud login works

SECTION 4 — Create GitHub Repository

GitHub acts like a cloud storage system for code.

Developers use it to:

• store projects
• collaborate
• deploy apps
• manage version history

Now let’s create a repository.

Step-by-Step Actions

Scene Checklist

Scene 1 — Open GitHub

Go to:

https://github.com

Scene 2 — Create Repository

Click:

• New Repository

Repository name:

deployable-ai-app

Select:

• Public

Click:

• Create Repository

SECTION 5 — Connect Local Project to GitHub

Now we connect our local project folder to GitHub using Git commands.

This allows GitHub to track and store our project files.

Terminal Commands

Open terminal inside project folder.

Run:

git init

Then:

git add .

Then:

git commit -m “Initial commit”

Now connect GitHub repository:

git remote add origin YOUR_GITHUB_REPOSITORY_URL

Push files:

git branch -M main

git push -u origin main

Notice how GitHub shows an error? It is because it has detected the API key.

SECTION 6 — Important Security Rule

Never upload your secrets file to GitHub.

Your Gemini API key must remain private.

Let’s fix this!

Create a file named:

.gitignore

Add:

.streamlit/secrets.toml

This prevents sensitive files from being uploaded publicly.

SECTION 7 — Deploy Using Streamlit Community Cloud

Now we will deploy the app publicly using Streamlit Community Cloud.

This platform connects directly to GitHub.

Whenever your GitHub code changes, the app updates automatically.

Step-by-Step Actions

Go to:

https://share.streamlit.io

Deployment Steps

1. Sign in with GitHub
2. Click:
   • New App
3. Select repository:
   • deployable-ai-app
4. Select branch:
   • main
5. Select file:
   • app.py
6. Click:
   • Deploy

It shows that the API key is missing. That’s because our deployed app still needs the Gemini API key.

We need to securely add it inside Streamlit Cloud secrets.

SECTION 8 — Add Gemini API Key to Streamlit Cloud
Steps

Inside Streamlit Cloud:

• open App Settings
• open Secrets

Paste:

GEMINI_API_KEY = “YOUR_API_KEY”

Save.

Reboot app.

SECTION 9 — Final Deployment Demo

Your app is now live on the internet.

Anyone with the link can use your AI application.

This is a major milestone in your AI development journey.

Demo Prompts

What is generative AI?

Then:

Explain it like I am a beginner.

SECTION 10 — Real Product Perspective

Most modern AI startups use workflows very similar to this:

• GitHub for source control
• cloud deployment platforms
• API-based AI models
• public web interfaces

You are now following a real industry workflow.

SECTION 11 — Lesson Recap

Recap Slide

You learned how to:

Create a GitHub repository

Push code to GitHub

Protect API keys

Deploy a Streamlit app

Create a public AI application