1. Create your track folder
mkdir -p ~/dev/analytics
cd ~/dev/analytics
2. Analytics tools: let Claude Code do it
Open Claude Code in your track folder:
claude
I'm setting up an analytics and BI environment. Please:
1. Install Python 3.11+ via Miniconda, then create a conda environment called "analytics"
2. Install core packages in the analytics environment: pandas, jupyter, matplotlib,
seaborn, plotly, scipy, statsmodels, duckdb
3. Check if Docker is installed. If not, tell me how to install it (it needs admin access)
After each step, verify it worked and show me the result.
Verify
Once Claude Code finishes:
conda activate analytics
python --version
python -c "import pandas; import matplotlib; import plotly; import scipy; import duckdb; print('All packages installed')"
jupyter notebook --version
You should see Python 3.11+, "All packages installed", and a Jupyter version number.
3. Your first look
Everything is installed. Before you start Project 1, see what Claude Code can do when you point it at an analytics problem.
Create a small CSV dataset of 500 e-commerce orders with columns: order_date,
product_category, revenue, customer_segment, region, is_returned. Then load it
into DuckDB, write SQL queries to find: total revenue by category, return rate
by segment, month-over-month revenue trend. Produce 3 clean charts that a
stakeholder could understand at a glance. Write a brief executive summary of
the key findings.
As you work through the track, you'll learn why a single prompt isn't enough: why that revenue calculation might use the wrong definition, why those charts might be misleading, why "return rate by segment" might hide a Simpson's paradox, and why a stakeholder would challenge your conclusions.
But for now, look at what just happened. That's the starting point.