Artificial Intelligence Integrated Development

AI Integrated Automation Developments - Advance

Theme: Development With AI — Build, Deploy and Use Your Own Automation Tool

Audience: Smaller group — participants who are comfortable with Basic and intermediate and have technical interest

Total Duration: 3 Hours (2 sessions × 90 minutes)

Format: Presentation + Guided Development + Group Demo

Goal: Advanced coding-assisted development — participants build, deploy and demonstrate a real working AI tool

Session 3A — Theory and Presentation | 90 Minutes

Session Title

AI-Assisted Development: Build Your Own Tool — GitHub, Hugging Face Hosting and Deployment

Session Objectives

By the end of this session, participants will be able to:

1. Explain what AI-assisted coding is and why it does not require a programming background
2. Set up a GitHub account and understand its purpose — repository, commit, version control and deploy
3. Understand Hugging Face as a hosting and deployment platform for AI models and apps
4. Use an AI-assisted coding tool (Cursor or Replit) to generate code from plain-English instructions
5. Describe what an AI agent is and name key agent libraries: LangChain, AutoGen, CrewAI
6. Read and edit a Python script by describing changes to AI
7. Identify free deployment options: Replit, Vercel, Render and Hugging Face Spaces

Content to Be Taught

1. AI-Assisted Coding — No Programming Background Needed
   1. Describe what you want in plain English — AI writes the code
   2. Tools: Cursor, Replit, GitHub Copilot
   3. How non-developers can build working tools with AI assistance

1. GitHub Basics for Beginners
   1. What GitHub is: a place to store, version and share code
   2. Repository, commit, branch and deploy explained simply
   3. Why version control matters — never lose your work
   4. Account setup walkthrough and first repository

1. Hugging Face as a Hosting Platform
   1. Hugging Face Spaces: deploy a web app or AI model with a live URL — free
   2. How to push code from GitHub to Hugging Face Spaces
   3. Gradio and Streamlit: simple Python frameworks for building AI app interfaces
   4. Why Hugging Face is ideal for sharing AI tools with colleagues and the public
   5. Comparison: Hugging Face Spaces vs Replit vs Vercel vs Render

1. AI Agents Explained
   1. What an agent is: AI that takes actions, not just answers
   2. LangChain, AutoGen, CrewAI — overview of agent libraries
   3. Example: an agent that reads email, decides action and drafts a reply
   4. Multi-agent systems: AI teams working together on a task

1. Python Basics With AI Help
   1. You do not need to memorise Python — AI writes it
   2. Reading a script well enough to understand what it does
   3. Editing a script by describing changes to AI in plain English
   4. Understanding API keys: what they are and how to use them safely

Session Outcome

Outcome: Participants understand that AI has removed the main barrier to building tools — no programming background required. They know how to use GitHub for version control and Hugging Face for free deployment, and are ready to attempt their first guided build in Session 3B.

Session 3B — Practical / Build and Deploy  |  90 Minutes

Session Title

Build, Deploy and Demo: Your First AI-Powered Office Tool

Session Objectives

By the end of this session, participants will be able to:

1. Use Replit or Cursor with AI assistance to generate a working script from a plain-English description
2. Build one of the two guided projects: Office Letter Generator or Attendance Email Sender
3. Push code to GitHub and deploy the tool to a live web link using Hugging Face Spaces, Replit or Vercel
4. Present and demonstrate the working tool to the group
5. Describe a personal AI action plan for their role — one automation to implement this month

Build and Deploy Steps

1. Setup (15 minutes)
   1. Participants choose one of the two build options below
   2. Open Replit or Cursor, log in with GitHub account
   3. Facilitator demonstrates: describe the tool to AI in plain English, generate the first code, review it together

1. Build Option A: Office Letter Generator (45 minutes)
   1. A simple web form where the user enters name, subject and purpose
   2. The form sends a prompt to the AI API and returns a formatted letter
   3. Participant uses Cursor or Replit with AI to generate and adjust the code

1. Build Option B: Attendance Email Sender (45 minutes)
   1. A script that reads a list of names and sends a reminder email to each
   2. Uses a free email API: Resend or Gmail SMTP
   3. Participant uses AI to generate, test and adjust the script

1. Deploy and Share (15 minutes)
   1. Push code to GitHub repository
   2. Deploy to Hugging Face Spaces, Replit or Vercel — choose one
   3. Each participant shares their live tool link with the group

1. Group Demo and Closing Discussion (15 minutes)
   1. Each participant demonstrates their tool — two minutes each
   2. Peer feedback: what worked, what to improve
   3. Closing discussion: personal AI action plan — one AI task to automate this month
   4. Data privacy policy and AI governance for council staff
   5. Resource pack handout
   6. Q and A open floor
   7. Certificate presentation and programme close

Session Outcome

Outcome: Participants build and deploy a real working AI-powered tool. They leave with a live URL, a GitHub repository, a personal AI action plan and the confidence to continue learning and building independently.