Discovery
Start with incidents, toil, data, controls and business constraints—not a predetermined agent.
MODULE 10 / 10
All modules ↑VALIDATE · TARGET L4 · 7–10 days
10
AI-SRE readiness audit
Deliver a reviewed readiness assessment and roadmap to one warm design partner without claiming unproven automation value.
Package internal learning into a bounded client assessment that is honest about evidence, access and uncertainty.
Prerequisites
- Modules 01–09 gates passed by pod
- Internal pilot metrics reviewed
- Security assessment rehearsed
01 / CONCEPT LESSONS
Build the mental model first.
Maturity model
Assess foundations, telemetry, runbooks, identity, safety, economics and operating ownership.
Opportunity selection
Rank use cases by value, evidence availability, action risk and implementation effort.
Architecture options
Offer assistive, approval-based and bounded automation options with explicit trade-offs.
Roadmap and proof
Recommend a smallest safe experiment, success measures, stop criteria and responsible owners.
02 / SYSTEM FLOW
Trace the complete path.
Sponsor goals→Technical discovery→Evidence review→Maturity score→Opportunity ranking→Architecture options→Pilot roadmap
For every arrow, the engineer must name the data, identity, trust boundary, failure mode and evidence produced.
03 / GUIDED LAB
Build it in bounded steps.
AI and documentation are allowed. The learner owns every generated artifact and must preserve the stated safety boundaries.
- 01Run an internal mock discovery.
- 02Collect evidence for every score.
- 03Rank three use cases.
- 04Design one smallest safe pilot.
- 05Estimate measurement and operating cost.
- 06Present findings to a hostile review panel.
04 / INCIDENT SIMULATION
The client cannot provide production access or reliable incident data, but expects ROI numbers.
Mentor injects
- Sponsor asks for guaranteed savings
- Runbooks are incomplete
- Security will not approve write access
Required evidence
- Assumption register
- Reframed read-only discovery plan
- Range-based estimate and validation gate
Investigation loop
Symptom → impact → hypothesis → evidence → test → root cause → containment → correction → verification.
05 / DESIGN CHALLENGE
Start with a blank page.
Compare no-agent process improvement, investigation copilot and approval-based agent for the client’s top use case.
06 / VERIFY & SUBMIT
Verification checklist
Submission package
- Discovery notes
- Maturity assessment
- Opportunity matrix
- Architecture options
- 90-day roadmap
- Executive presentation
Review questions
- What problem is worth solving?
- Which evidence is missing?
- Why an agent?
- What is the safest first autonomy level?
- How will the client know to stop?
07 / MENTOR-ONLY GUIDANCE
Reward honest uncertainty. Fail any assessment that converts assumptions into precise ROI or recommends autonomy without operational foundations.
SCORING1 · cannot explain2 · explains with gaps3 · implements and verifies4 · designs and handles trade-offs
Record the achieved L-level only from observed evidence. Course completion and certification do not automatically change the skill matrix.