Analytics Immersive · syllabus
Twelve weeks, twelve things you have to hand in.
This is the flagship calendar, published in full. Engineering students add a four-week operations block; ML evenings follow the same spine at half speed. Nothing here is a teaser — the reading lists and rubrics go out with your acceptance letter.
01 · Foundations
Weeks 1–3
02 · Measurement
Weeks 4–6
03 · Craft
Weeks 7–9
04 · Capstone
Weeks 10–12
Week 1
The shape of data
Types, tables, joins, and the query plan. We start on a whiteboard before a keyboard.
Hand-inER sketch of a live public dataset
Week 2
SQL, properly
Window functions, CTEs, and why your GROUP BY is lying to you.
Hand-in10-query audit of a messy retail database
Week 3
Python for analysts
pandas as a thinking tool: reshape, join, aggregate, repeat.
Hand-inNotebook cleaned to reviewer standard
Week 4
Statistics you can defend
Distributions, sampling error, confidence, and the hypothesis you should have tested.
Hand-inA/B readout memo with caveats
Week 5
Experimental design
Power, randomisation, and the metrics that look good but mean nothing.
Hand-inDesign doc for a realistic experiment
Week 6
Mid-term review
Panel format: defend an analysis in front of mentors and peers, cold.
Hand-in15-minute defence, scored rubric
Week 7
Visual encoding
Data-ink, colour for meaning, and the chart types that deserve extinction.
Hand-inRedesign of a real corporate chart
Week 8
Dashboards people read
Layout hierarchy, filters with memory, and the one-number headline.
Hand-inExecutive dashboard, one screen
Week 9
Narrative
The memo, the slide, the sentence. Writing that survives being forwarded.
Hand-inFindings memo to a fictional board
Week 10
Capstone sprint
Real partner data, real stakeholders, daily standups. Faculty as consulting desk.
Hand-inPartner-approved analysis plan
Week 11
Build week
Pipeline, analysis, dashboard, and documentation as one deliverable.
Hand-inWorking capstone repo
Week 12
Hiring demo day
Ten minutes on stage, then tables with hiring partners. Portfolio reviewed live.
Hand-inDemo + reviewed portfolio
The defence rubric
Every graded hand-in is scored on four axes: correctness of method, honesty of caveats, clarity of the written memo, and reproducibility of the repo. Two axes below the line means a rewrite, not a fail — but you do it on your own time.
Contact hours
22 hours in class per week, plus open lab nights on Thursdays. Faculty hold two office-hour blocks daily; the median student uses one and a half and graduates anyway. The reading is roughly eight hours more. It is not optional.
What we don't teach
Deep learning, LLM fine-tuning, and anything we cannot staff with a practitioner who does it for a living. When those earn their place in a working analyst's week, they'll appear in this syllabus — dated, with the faculty member who teaches it.
Evenings are part of the course.
The lab stays open until nine. Cohorts 8 through 11 logged a combined 4,300 lab hours — most of them after the lectures ended and before the demo-day panic properly started. Bring a problem; the whiteboards are always free.
Admissions & costs