Interview Prep — Rack Management (Senior / Staff) SWE

This is the interview surface for the role, drawn from the 13 phases. A Qualcomm-style rack-management loop spans deep technical (systems programming, protocols, hardware), applied infra (provisioning, orchestration, observability), production debugging/RCA (the most-weighted theme), system design, and behavioral/Staff leadership — often in the same day.

The five question banks

How to use this

1. Self-test cold, then check against the model answer and the WARMUP it derives from. A gap points you back to a phase. The WARMUPs' "Interview Q&A" sections are the depth target — this bank is the index + the highest-yield additions.
2. Derive, don't recite: every technical answer should reason from a principle (the roofline/ bandwidth math, the 80% breaker rule, follow-the-link Redfish, level-triggered reconcile, verify-before-and-after firmware), not a memorized fact.
3. Debugging is the headline: the JD repeats "debugging and root-cause analysis of complex issues, including deployed systems" three times. Be ready to narrate method (Phase 10) on any "this is broken in production" prompt.
4. Hands-on credibility: the labs are your evidence. "How do you test code that needs a BMC?" or "how do you parse a crash dump?" should be answered with something you built.

The 20 questions most likely to decide the loop

1. Walk me through the components of an AI rack and how it's powered and cooled. (P01)
2. Two 30A 415V feeds, N+1 — how much load can you safely place, and why not 100%? (P01)
3. What is Redfish and why is the industry moving to it from IPMI? (P03)
4. Discover a server's power state and CPU temperature with Redfish — walk me through it. (P03)
5. How does a firmware update work safely, and what are A/B banks? (P08)
6. What's PCIe AER? Correctable vs uncorrectable — what do you do about each? (P04/P08)
7. A node logs a flood of correctable ECC errors — is that a problem? (P08)
8. What is RAII and why does it matter for a rack daemon? (P02)
9. subprocess.run(..., shell=True) — what's wrong and how do you fix it? (P02)
10. How do you test software that talks to hardware you don't have in CI? (P11)
11. Explain the operator pattern and the reconcile loop. Why level-triggered? (P06)
12. Walk me through what happens when a diskless server PXE-boots. (P05)
13. What is idempotency and why is it central to config management? (P05)
14. Design a telemetry pipeline for a rack fleet with Redfish/IPMI/SNMP devices. (P07)
15. What would you alert on for an AI rack, and how do you avoid alert fatigue? (P07)
16. A production node reboots intermittently under load — debug it, you can't take it offline. (P10)
17. What's a circuit breaker and why do you need one with retries? (P10)
18. How do you isolate tenants on a shared rack, and why is the management plane special? (P09)
19. Secure boot vs measured boot — what's the difference, and what's attestation for? (P09)
20. What's the difference between Senior and Staff, and how do you drive implementation/mentor? (P12)

Each maps to a WARMUP's Interview Q&A with a full principal-level answer.

Format expectations for this role

• Technical screens: reason about a roofline/power budget; whiteboard a Redfish walk, a reconcile loop, or a firmware state machine; read a backtrace / RAS log.
• Coding: a daemon/CLI/parser in Python or C++ (the labs are calibrated to this) — with the seam, error handling, and tests.
• System design: the system-design walkthroughs (control plane, telemetry, fleet firmware, tenancy, provisioning).
• Debugging: a "production is broken" scenario — narrate the method (Phase 10).
• Behavioral/Staff: STAR stories, design-review judgment, distributed-team communication.