# AI Roadmap Explorer — Quest Path & Class Dashboard (Streamlit + Plotly)

## PURPOSE
The video ends with a summary slide of every quest path and, separately, a slide per
class. Nobody can hold eight courses times six classes in their head. This dashboard puts
the whole map on one screen: a scatter of quest paths by study hours against monthly cost
so the price of each route is obvious, a bar chart of how long a working skill tree takes
per class, and a Tutorial Island checklist of the starter pack.

## WHY THIS MATTERS
It visualizes the transcript's central claim — that AI learning is a map with a shared
onboarding and divergent classes — and its practical constraint, the budget. The video
repeatedly ties cost to choice: the ~$49/month Coursera Plus subscription (7-day trial)
behind Google Essentials and the Google AI Professional Certificate, Andrew Ng's videos
free to audit with Pro labs at ~$25–30/month, and the genuinely free paths from
Elements of AI, Anthropic, OpenAI, Kaggle, and Microsoft/LinkedIn. Seeing hours plotted
against dollars turns "paradox of choice" into a decision.

## HOW TO BUILD
Stack: **Streamlit + Plotly + pandas**. Keep two small DataFrames (`QUESTS`, `CLASSES`),
build the figures with `plotly.express`, and let Streamlit's sidebar widgets filter the
frame before every render.

```python
view = QUESTS[(QUESTS["usd_month"] <= budget) &
              (QUESTS["style"].isin(styles))]
st.plotly_chart(px.scatter(view, x="hours", y="usd_month",
                           color="style", size="hours", text="path"),
                use_container_width=True)
```

A `st.checkbox` loop renders the three starter-pack skills as a trackable list.

## WHEN TO USE IN PRODUCTION
Use it as the landing/demo artifact for a course marketplace, an academy open-house, or a
"choose your path" page where stakeholders need to see coverage and cost at a glance. It
is ideal for content and marketing teams because it is data-only and deploys to Streamlit
Community Cloud in minutes. Do NOT use it for transactional flows or auth-gated personal
dashboards (Streamlit is not an app framework for that), nor for very large datasets that
belong in a real BI tool.

## HOW IT HELPS STUDENTS
Students learn declarative charting, reactive filtering, and how to tell a data story —
the visualization half of production readiness that almost every ML curriculum skips.

## STARTER CODE
`starter/main.py` ships the two DataFrames, the scatter and bar builders, the sidebar
filters, and the starter-pack checklist, with `# TODO` markers for live CSV loading and
progress persistence.
