Advanced · 5 modules
On-Call — Incident Command for Platform Engineers
Run calm incidents, write blameless postmortems, ship reliability
A 4-week cohort with a live game-day. You'll command a simulated outage, coordinate responders, communicate to stakeholders and write the postmortem.
- Level
- Advanced
- Length
- 4-week cohort + live drills · 5 modules
- Topics
- 16
- Language
- EN
Marek Lis
Priya Raman
Program
What you will do
5 lessons · 16 topics · about 8 h of material
- MOD 0031 min
Briefing
Cohort welcome, schedule, grading and the rules everyone follows on the bridge.
VideoReading
- MOD 011 h 25 min
Signals
Read dashboards like an IC and alert on what users feel.
ImageReadingEmbed
- MOD 0257 min
Command
Take command, keep the bridge disciplined and decide under pressure.
VideoAudioInteractivePDF
- MOD 032 h 45 min
Game day
Command a simulated cascading failure, run the live drill and check your comms.
SimulationInteractiveTracked activityQuiz
- MOD 042 h 45 min
After
Turn the outage into learning: blameless postmortem, review, reading.
ReadingProject
Stay calm at 3 a.m. On-Call is a four-week cohort for engineers who carry the pager, or are about to. You will learn to take command of an incident, coordinate responders, keep stakeholders informed and turn every outage into a better system. In week three you run a live game day: a simulated cascading failure in a sandbox cluster, graded on how you command it.
Curriculum
| Module | Week | Formats |
|---|---|---|
| 0. Briefing | 1 | kickoff video, rules of engagement |
| 1. Signals | 1 | annotated dashboard, SLOs and error budgets, conference talk |
| 2. Command | 2 | SEV1 role-play video, bridge-call recording, branching scenario, printable runbook |
| 3. Game day | 3 | SCORM outage simulator, xAPI-tracked live drill, comms quiz |
| 4. After | 4 | blameless postmortems, postmortem assignment with peer and tutor review |
Outcomes
- Declare and classify incidents with a shared severity matrix
- Run the first five minutes of a SEV1: roles, channel, first update
- Communicate impact, current action and next update time to any audience
- Alert on SLO burn rate instead of on causes
- Write a blameless postmortem with owned, dated action items
Format
Four weeks, about 4–5 hours a week, plus a 90-minute live drill. Each cohort has at most 24 engineers, so every drill gets individual feedback. Deadlines are real: the comms quiz closes on Thursday of week 3, the postmortem is due at the end of week 4.
Instructors
Priya Raman is a Staff SRE who has run the incident programme for a platform of 400 services, and wrote its runbook, severity matrix and postmortem template.
Marek Lis has been an incident commander at a fintech processing card payments in 30 countries. He runs the game day and grades the drills.
Complete all modules and pass the drill to earn Incident Commander — Level 1.