Instructor Setup Checklist

Work through this at least one week before Day 1. Most items only need doing once.

Before the course starts

Software (you, the instructor)

• Excel — Microsoft 365 (cloud) or Excel 2021+ on Windows/Mac. Confirm XLOOKUP is available (older versions only have VLOOKUP).
• SQLite — already on macOS/Linux. On Windows: download from sqlite.org/download.html (the “sqlite-tools” zip — extract sqlite3.exe somewhere on PATH). Test: sqlite3 :memory: "SELECT 42".
• DB Browser for SQLite — free GUI from sqlitebrowser.org. This is what students will use day-to-day.
• Python + Jupyter — Anaconda or pip install jupyterlab pandas matplotlib seaborn. Test: jupyter lab.
• Power BI Desktop — free, Windows only. Mac users: use a VM or the web service.
• Claude Code — sign in at claude.ai/code and confirm you can launch a session.

Software (each student)

Send students this list one week ahead. Day 1 morning runs much smoother if installs are done.

• Excel (any modern version with XLOOKUP)
• DB Browser for SQLite (free GUI) — for Day 2. SQLite itself is already on Mac/Linux; Windows users grab sqlite3.exe from sqlite.org/download.html.
• Anaconda or an account at JupyterLite (in-browser, zero install) — for Day 3
• Power BI Desktop (Windows) — for Day 4
• An account at claude.ai — for Day 5

Dataset

• Run bash data/olist/download.sh and confirm all 8 CSVs land in data/olist/.
• Open olist_orders_dataset.csv in Excel — confirm it opens in under 5 seconds.
• Build the SQLite database: cd data/olist && sqlite3 olist.db < load_into_sqlite.sql. Should print row counts at the end.
• Sanity-check: sqlite3 data/olist/olist.db "SELECT COUNT(*) FROM orders" → should print 99441.

Capstone dry run

• Spend 90 minutes doing the full capstone yourself, end-to-end, using only the tools students will have. Note anywhere you got stuck — those are the spots students will get stuck.

Logistics

• Confirm screen sharing works in your conferencing tool — students need to see DAX formulas and Python code clearly.
• Set up a shared chat channel (Slack/Teams/Discord) for between-session questions.
• Decide on a recording policy and tell students Day 1.

Day-by-day pre-class checks

Before Day 1 (Excel)

• Walk through docs/curriculum/day1_excel/01_tables.md end-to-end in your Excel and confirm every “Try it yourself” drill runs in the time budget the lesson states.
• Pre-download Olist CSVs to a USB stick or shared drive in case student internet is patchy.

Before Day 2 (SQL)

• Verify every student can open data/olist/olist.db in DB Browser for SQLite. If a student’s olist.db doesn’t exist, walk them through cd data/olist && sqlite3 olist.db < load_into_sqlite.sql.
• Have a pre-built olist.db on a USB stick or shared drive as a fallback for students whose CSV download is broken.

Before Day 3 (Python)

• Decide: are students using local Jupyter or JupyterLite (browser)? Pick one and stick with it for class consistency.

Before Day 4 (Power BI)

• Mac users: confirm their VM/web access works. Power BI Desktop is Windows-only — don’t let this surprise anyone Thursday morning.

Before Day 5 (Claude Code)

• Confirm every student can sign in at claude.ai/code.
• Have a fallback: if claude.ai is unreachable from corporate network, use the desktop app or CLI.