Instantiation of generate-exam.md for the standard midterm. Read that file for the LaTeX conventions and workflow; use these parameters.

Parameter	Value
Exam type	midterm
Agenda	docs/agenda/<term>.md, Meetings 1–8 (through Model Training and Evaluation), or whatever range the agenda says the midterm covers
Past exams	docs/midterm/<YYYY>/ — the two or three most recent. From 2026 on, each also carries rubric.yaml (how it was scored), grading-notes.md (decisions made while giving it), and analysis.md (where the class split from the key): read those three first, and do not reuse a question they mark as ambiguous
Length	4 pages, 50 points, designed for 30–40 minutes; ends with a 3-point feedback section
Must include	one question on Assignment 1; questions on the hands-ons the agenda records in the meeting range; every midterm flag in the range
Instructor output	/Users/feamster/Dropbox/Tmp/exams/ml-systems/midterm/<YYYY>/ (not committed until after the exam; when the instructor publishes it to docs/midterm/<YYYY>/, rubric.yaml, grading-notes.md, and analysis.md from the grading directory in instructor-data go with it, never the scores or review pages)

Follow the agenda for what to cover. As orientation only, midterms usually draw on: motivating applications (QoE, security, resource allocation); passive vs. active measurement and flow records; feature engineering and representation; data quality; the pipeline (splits, cross-validation, leakage); evaluation metrics; and bias–variance. Include a topic only if that term’s agenda covers it in the range (drift, for example, is often taught later in the term) and never one the agenda says will not be asked.