Module 1 cheat sheet: Introduction to the Program
The big picture of AI, the jobs it creates, your first deployed GenAI app, and a clean working setup.
Key ideas
- AI ⊃ Machine Learning ⊃ Deep Learning ⊃ Generative AI. S1 →
- ML learns rules from data; traditional code follows rules a person wrote. S1 →
- NLP understands language, GenAI creates content, agents take actions with tools. S1 →
- A GenAI app can be ~30 lines: Streamlit UI + one LLM API call. S2 →
- One virtual environment per project; keys in .env, never in code. S3 →
Job roles
| Role | Focus |
|---|---|
| GenAI Engineer | LLM features: chatbots, RAG, summarisation |
| LLM Engineer | fine-tuning, serving, optimisation |
| AI Product Engineer | end-to-end AI products and UX |
| LLMOps Engineer | deploy, monitor, tracing, cost |
Everyday commands
python -m venv .venv && source .venv/bin/activate # Windows: .venv\Scripts\activate pip install -r requirements.txt git checkout -b my-feature git add . && git commit -m "message" && git push streamlit run app.py
Streamlit vs Gradio
| Streamlit | Gradio | |
|---|---|---|
| Style | script reruns top to bottom | wrap a function |
| Best for | dashboards, data apps | quick model demos, chat |
| Free hosting | Streamlit Community Cloud | Hugging Face Spaces |
Common mistakes
- Committing an API key or .env file to GitHub.
- Installing everything into one global Python.
- Deploying a public app with no spending limit on the API key.
- Calling every chatbot an “agent”: agents use tools and act.
Created by Ajeetkumar · EasyLearnSpace