The manager's case
One engineer's packet, drawn by lottery in week 8: the case you would argue for them in the room, and what in the packet you do not believe.
Due: 7 May 2027 · Weight: 20%
The brief#
You are the manager. Here is the packet. Write the two minutes. Then write what you do not believe.
The old manager moved teams in the April reorg. You have inherited one of the twelve engineers on the team page, and in week 11 you will sit in a calibration meeting and argue for them. You have never met them. This is normal; week 9 told you why.
The packet#
Six parts, the same for every engineer. Read them in this order the first time: the numbers, then where they sit, then the announcement, then the peers, then the manager’s notes, and the self-review last. Then read the self-review again with the rest in mind. Most packets have something in them that this course has a week for. Some have a person in them the numbers cannot see. Deciding which is the job.
What to write#
Page one: the case. 450 words or fewer, because you will read it aloud in two minutes and the chair will stop you at two. It opens with one sentence: the rating you propose, and the curve the room will be working to (one in five exceeds, one in ten below). Then three claims. Each claim points at one item in the packet and names, in brackets, what kind of evidence it is: self-review, peer, numbers, trace, or manager’s note. No “I believe”; only “the packet shows”. Write it as the manager who wants this person rated well, because in the room that is your job.
Page two: what you do not believe. A list. Every item in the packet that you think is a trick from this course, with the week it belongs to and what you would need in order to check it: a log, a commit history, a second peer, a number’s source. Be as hard on the packet as you were soft on page one. Then one closing paragraph: the rating you would give if everything on this page checked out the way you suspect, the gap between that and page one, and why you proposed page one’s rating anyway. That paragraph is the point of the assessment. Managers hold both ratings in their head; the course wants the second one on paper.
Rules#
- The case is for the engineer you were assigned. Swaps are not available; neither are they in companies.
- Page one stays private until the room. Do not share it with your group.
- Draft it however you like, with whatever you like. The mark is for whether the two pages say something true about the packet.
Timeline#
- Thursday 29 April, week 8. Groups drawn in the stand-up. Packets assigned by lottery within the group, one engineer each.
- Friday 7 May, 5 pm. Both pages due.
- Friday 14 May. Marks and comments back, so you have them before the room.
- Thursday 20 May, 2 pm. The room. See Assessment 3.
What you submit#
One PDF of two pages, named SLOP3385-<uid>-case.pdf, in the Assessment
2 thread of the course forum. The policies page says what counts as
submitted and how to get an extension.
Marker’s checklist#
- Is this the assigned engineer, and does page one open with a rating and the curve?
- Do the three claims each point at a packet item, and is the evidence type named correctly?
- Would page one take two minutes to read aloud?
- Does page two find what the packet holds, by week, without calling a real result a trick?
- Is each check on page two something a manager could actually do?
- Does the closing paragraph give a second rating, the gap, and a reason for still proposing the first?
The spec
The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.
- the case is for the engineer you were assigned, and page one opens with a proposed rating and the curve
- page one has three claims, each pointing at one item in the packet and naming its evidence type
- page one is 450 words or fewer
- page two names every item you do not believe, by week, with what it would take to check it
- page two ends with the rating you would give if everything checked out, and the gap
How it is marked
| Criterion | Weight |
|---|---|
| The proposed rating is argued from the packet, each claim tied to a named source | 35% |
| The tricks in the packet are found and named by week | 30% |
| What it would take to check each one is specific and checkable | 15% |
| Page one reads in two minutes, and page two is honest about the gap | 20% |
What each grade looks like
| Grade | Marks | What the marker sees |
|---|---|---|
| HD | 80–100 | Every claim on page one points at one item in the packet and says which kind of evidence it is. Page two finds every trick the packet holds, names each by week, does not mistake a real achievement for one, and says exactly what would be needed to check it. The last paragraph states the gap between the two ratings and why the first was still proposed. |
| D | 70–79 | Page one argues from the packet with one claim that leans on the self-review without saying so. Page two finds most of the tricks and names them correctly; one check is vague. The gap is stated but not explained. |
| CR | 60–69 | Page one is a case, but two of its three claims come from the same source. Page two finds the obvious trick and misses or misnames another, or calls a genuine result a trick. The gap paragraph gives the same rating twice without saying why. |
| P | 50–59 | Page one restates the self-review. Page two lists doubts without weeks or without saying what would resolve them. There is a rating on each page and no account of the difference. |
| N | 0–49 | Page one is over 450 words or could not be read in two minutes. Page two is missing, empty, or calls everything a trick. Or the case is for the wrong engineer. |