🤖

Introduction to AI in Development

Sharing Session - Week 1

AI Fundamentals ML Basics Tools Company Direction

2h

Duration

4

Topics

All

Audience

SLIDE 02 / 18

🎯 Objectives

Session Objectives

By the end of this session, you will:

🧠

Understand AI & ML

Basic concepts and how they differ

🛠️

Know the Tools

AI tools available in the market

🎯

Company Direction

How we're adopting AI as a tool

🚀

Next Steps

How to get started with AI assistance

SLIDE 03 / 18

🤖 Basics

What is Artificial Intelligence?

AI is:

Systems that can perform tasks requiring human intelligence
Pattern recognition at scale
Learning from data
Making predictions or decisions

🧠

Key Insight

AI doesn't "think" like humans — it finds patterns in data

AI Capability Evolution

SLIDE 04 / 18

📚 Definitions

AI vs ML vs LLM

🤖

Artificial Intelligence

Broad field of computer systems performing intelligent tasks

Umbrella term

📊

Machine Learning

AI that learns patterns from data without explicit programming

Subset of AI

💬

Large Language Model

ML trained on text to understand and generate language

Subset of ML

Relationship

SLIDE 05 / 18

⚙️ How It Works

How LLMs Work (Simplified)

1. Training

Model learns patterns from billions of text documents

2. Input (Prompt)

You provide a question or task

3. Processing

Model predicts the most likely next words

4. Output

Generated response based on patterns learned

LLMs are prediction engines, not knowledge databases

SLIDE 06 / 18

💪 Strengths

What AI Does Well

✍️

Text Generation

Writing, summarizing, translating

💻

Code Assistance

Writing, explaining, debugging code

🔍

Pattern Recognition

Finding patterns in large datasets

🔄

Repetitive Tasks

Automating routine work

SLIDE 07 / 18

⚠️ Limitations

What AI Struggles With

🎯

Accuracy

Can produce confident but wrong answers ("hallucinations")

📅

Current Information

Knowledge has a cutoff date

🧮

Complex Reasoning

Multi-step logic can fail

🔐

Sensitive Data

Privacy and security concerns

Always verify AI output — trust but verify

SLIDE 08 / 18

🛠️ Tools

AI Tools in the Market

🟣

Claude

Anthropic

🟢

ChatGPT

OpenAI

🔵

Gemini

Google

⚫

Copilot

GitHub/Microsoft

🟠

Cursor

AI IDE

🔴

Codeium / IntelliSense

Code Assistant

SLIDE 09 / 18

💻 Dev Tools

AI Tools for Developers

Code Assistants

GitHub Copilot
Claude Code
Cursor AI
Codeium
Amazon CodeWhisperer

Chat-based AI

Claude (claude.ai)
ChatGPT
Gemini
Perplexity

Tool Comparison

SLIDE 10 / 18

🟣 Our Choice

Why Claude Code?

🛡️

Safety Focus

Built with responsible AI principles

💻

Code Understanding

Excellent at reading and explaining code

📝

Long Context

Can handle large codebases

🔧

Tool Integration

Works with IDE and CLI

SLIDE 11 / 18

🏢 Direction

Less Coding, Fix New Problem

Jensen Huang quote on less coding and solving new problems with AI

SLIDE 11 / 18

🏢 Direction

Less Coding, Fix New Problem

Purpose v Task

Our purpose — to solve our client problem.
The code task, let AI handle it.

SLIDE 11 / 18

🏢 Direction

Company Direction on AI

AI as a Tool, Not a Replacement

We're adopting AI to augment developer capabilities, not replace human judgment

✓

Approved Uses

Code assistance & review
Documentation
Learning & exploration
Testing support

✗

Boundaries

No production secrets in prompts
No customer data sharing
Human review required
No blind trust

SLIDE 12 / 18

📋 Approach

Our Implementation Approach

Phase 1: Education

Training sessions on AI fundamentals and tools (this series)

Phase 2: Pilot

Small team trials with Claude Code

Phase 3: Guidelines

Establish best practices and policies

Phase 4: Rollout

Wider adoption with support

SLIDE 13 / 18

📈 Benefits

Expected Benefits

30%

Faster Development

50%

Less Boilerplate

↑

Code Quality

↓

Rework Time

Focus on what matters — let AI handle the routine

SLIDE 14 / 18

⚖️ Responsibility

Developer Responsibilities

👀

Review All Output

Never blindly accept AI suggestions

🔐

Protect Data

Don't share sensitive information

📚

Keep Learning

AI assists, doesn't replace skills

🏷️

Own Your Work

You're accountable for the final code

SLIDE 15 / 18

🚀 Start

Getting Started

This Week

Create Claude account
Try simple prompts
Explore the interface

Coming Sessions

Week 2: Claude Code for dev work
Week 3: Agents & Skills
Week 4: CI/CD Integration

SLIDE 16 / 18

🎯 Summary

Key Takeaways

🧠

AI = Pattern Recognition

Not magic, not consciousness

🛠️

AI = Tool

Augments, doesn't replace

⚖️

Responsibility Remains

You own the final output

SLIDE 17 / 18

💬 Discussion

Open Discussion

What tasks would you like AI to help with?
What concerns do you have about AI tools?
How do you see AI changing your workflow?

SLIDE 18 / 18

🔮

Week 2 Preview

Coming Next

🐛

Bug Fixing with Claude

AI-assisted debugging

📐

Code Standards

Refactoring & quality

🧪

TDD Support

Test-driven development

🏠 All Sessions