Learn Python Data Analysis with Phoebe

Decode the data job market, learn real analysis

Eight 45-minute sessions, one messy real dataset: 124,000 LinkedIn job postings. You load it, clean it, interrogate it, chart it, and by Session 8 ship a report you'd be proud to post. NumPy, pandas, matplotlib, seaborn, and a first ML model - taught as the workflow analysts actually use. The hands-on sequel to learn-python-with-phoebe.

8sessions
124thousand real rows
45min per session
1shareable report shipped
🧭 Session 1 · easiest

Set up, and meet the mission

Get the analysis stack running - conda, Jupyter, pandas - load the 124k-row dataset, and take your first profiled look at the mess.

▶ Start hereA little Python assumed 🔢 Session 2 · easy

NumPy foundations

The fast numeric core under pandas: arrays, dtypes, vectorization, broadcasting, boolean masks - the mental model behind every quick operation.

Loops without loops 🐼 Session 3 · medium

pandas core: the DataFrame

Your home for the rest of the course. Series and DataFrames, loc vs iloc, boolean filtering, and answering your first real questions of the data.

Select, filter, summarize 🧹 Session 4 · getting real

Cleaning messy data

See the mess before you fix it: missingno maps, the 70% missing salary problem, dtype coercion, parsing salaries, tidying text, killing duplicates.

80% of the real job 🔗 Session 5 · getting real

Explore and aggregate

Where questions get answered: groupby split-apply-combine, pivot tables, and merging the separate tables into one picture of the market.

Findings, not tables 📊 Session 6 · medium

Matplotlib fundamentals

The engine under every Python chart. Figure and Axes anatomy, the four charts you need, labels and small multiples, saved at publication quality.

A chart is an argument 🎨 Session 7 · getting real

Seaborn and the polish

One-line statistical charts that speak DataFrames: distributions by group, correlation heatmaps, themes, and a cohesive publication-grade chart set.

Pretty, and correct 🚀 Session 8 · hardest

From insight to a shareable report

Turn seven sessions into findings people act on: five charts, five findings, a first salary model, and a LinkedIn-ready report with your name on it.

The capstone
easy - foundations medium - core skills getting real - the messy middle hardest - the capstone

Choose your path 🗺️

Everyone starts at Session 1. The sequence is the course - but the pace is yours.

🌱 The full journey (recommended) 1→ 2→ 3→ 4→ 5→ 6→ 7→ 8
⚡ Know Python, want pandas fast 1→ 3→ 4→ 5→ 8→ loop back for viz (6-7) when you need it
📈 Here for the visualization 1→ 3→ 6→ 7→ 8

The knowledge map 🧠

The whole course at a glance - hover a session to spotlight its concepts, click any node to jump in.