AI in Real Workflows
From Prompting to Agents
👋 Before vs After AI
❌ Before
- Manual coding
- Debugging alone
- Slow data exploration
- Repetitive tasks
✅ After
- AI-assisted coding
- Faster debugging
- Rapid prototyping
- Automation mindset
🚀 The Shift
I don’t code alone anymore.
I orchestrate AI.
🧠 How I Use AI
📊 Data Work
- SQL/NoSQL generation
- Data cleaning
- Analysis
🤖 AI Agents
- Natural language queries
- Retrieval + LLM
- Production systems
💻 AI as a Coding Agent
- Write functions
- Refactor code
- Generate tests
- Debug errors
Think: delegate, not write
⚡ Live Demo
Take messy code → improve it
- Refactor
- Add tests
- Explain logic
⚡ Live Demo
📊 AI for Data Work
- Prompt → SQL
- Prompt → Python
- Prompt → Insights
🤖 AI Agents (Simple View)
User Question → Search → Context → LLM → Answer
🔥 Why This Matters
- Works across industries
- Scales your productivity
- Reduces repetitive work
🧪 Hands-On Exercise
Build Your AI Assistant
🧩 The Problem
You have messy data / task
Your goal: - Extract insights - Automate something
🛠 Prompt Framework
- Role
- Task
- Context
- Constraints
- Output format
❌ Bad Prompt
Analyze this data
✅ Good Prompt
You are a data analyst. Analyze this dataset and return 3 key insights in bullet points. Focus on trends and anomalies.
🧪 Your Turn
- Pick a task from your field
- Use the framework
- Improve your prompt
🧠 Think Like This
How do I delegate this to AI?
⚙️ Optional Challenge
Build a Mini Agent
- Ask questions
- Refine answers
- Chain prompts
💬 Discussion
- How would you use this at work?
- What can you automate?
🚀 Final Thought
AI won’t replace you.
But people using AI will.