Roadmaps / Go / Interview topics
Reference
Interview topics
Use these for Go backend and platform roles. Each follow-up is the question a good interviewer asks next. The Staff line is what lifts the answer.
Phase 1
| Question | Follow-up | A Staff answer adds |
|---|---|---|
| How do you stop a goroutine leak? | How would you find one in production? | The leak profile, a goroutine count alert and a rule for every go |
| Channels or mutexes? | When is each wrong? | Ownership as the guide, with a measured case where one costs more |
| What does the Go scheduler do when a goroutine blocks? | What blocks a thread rather than a goroutine? | A way to detect it (the execution trace) and a fix |
| How do you limit concurrency per key? | How do you keep one slow key from blocking others? | Sizing with Little's law and a bound on each queue |
System design prompt (20 minutes): a webhook delivery service at 2,000 events a second. Outline: requirements and numbers (3 min), storage and claim (5), concurrency limits (5), failure cases (4), observability (3).
Phase 2
| Question | Follow-up | A Staff answer adds |
|---|---|---|
| How do you make delivery at-least-once safe? | What does the receiver need to do? | An idempotency key, a documented retry schedule and a replay tool |
| Which timeouts do you set on an HTTP server? | And on a client? | A budget per call that adds up inside the caller's deadline |
| How do two workers share a queue in Postgres? | What if a worker dies mid-job? | A lease with an expiry and a test that kills the worker |
| How do you design retries? | When do retries make things worse? | A budget, jitter and a breaker, tested against a recovering endpoint |
System design prompt (20 minutes): add a second region to the delivery service. Outline: what must be consistent (4 min), failover (5), duplicates (5), cost (3), rollout (3).
Phase 3
| Question | Follow-up | A Staff answer adds |
|---|---|---|
| A service's p99 doubled after a release. What now? | How do you decide to roll back? | A rule agreed before the release and a diff of metrics |
| How do you set a service level objective? | What does the error budget change? | A burn-rate alert and a rule for freezing risky releases |
| How do you find a memory growth? | What if you cannot take a heap profile? | Allocation profiles, sampling and a safe way to capture one |
| How do you test code that uses time? | Why not sleep? | synctest or an injected clock, and the tests that stay fast |
System design prompt (20 minutes): a Go API's p99 doubled after a release. Outline: first questions and rollback (3 min), metrics to compare (5), profiling plan (5), likely causes (4), the fix and the guard (3).
Phase 4
| Question | Follow-up | A Staff answer adds |
|---|---|---|
| A dependency has a critical flaw. What now? | How do you know if you are affected? | govulncheck, an inventory that answers in minutes and a patch SLA |
| How do you make builds reproducible? | Why does it matter? | A pinned toolchain, a check that two builds match and a signed release |
| How do you adopt a new Go release across teams? | What if a team cannot move? | A trial in one service, a rollout schedule and a rule for exceptions |
| How do you structure a service for many teams? | What do you standardise? | A template with logging, metrics, shutdown and lint set up, plus an opt-out |
System design prompt (20 minutes): a paved road for 50 Go services. Outline: what is standard and what is free (4 min), the template and its checks (5), upgrades (4), exceptions (4), measuring adoption and incidents (3).