Skip to content
InfraDrill

Distributed systems in a Senior backend interview

Every service a Senior designs is several machines pretending to be one. The round tests what that pretence costs — less about consensus algorithms than about what your system does while one of its parts is unreachable and nobody knows it yet.

Check whether this one is a gapTry real questions first — no signup

The check comes with every pass, covers all 8 competencies of a Senior backend loop, and takes 10–15 minutes. It reports each one as Ready, Borderline or a Gap for this level. It never predicts a pass.

Where this usually breaks down at Senior

These are the distinctions the questions on this axis are written to separate — our judgement from writing them and having them reviewed, not a measurement of candidates. We have run no study, so nothing here is a statistic and none of it is phrased as one.

  1. 1.

    At-least-once delivery without idempotency

    Nearly every real pipeline redelivers. The design question is not how to prevent the duplicate — in general you cannot — but what makes processing it twice harmless, where the deduplication key lives, and how long it is honoured.

  2. 2.

    Consistency named rather than chosen

    “Eventually consistent” is a property with a window. How long it is, what a reader sees inside it, and which operations cannot tolerate it are the actual answer; the label on its own commits to nothing.

  3. 3.

    Retries with no budget

    Retries compose down a call chain: three layers each retrying three times is not three attempts. Without a bound and a backoff, a dependency that merely got slow becomes an outage caused by its callers.

  4. 4.

    Partial failure treated as failure

    The hard case is not the dependency being down. It is three of five writes having landed and the process coordinating them having died. What happens to the other two, and who finds out, is the question.

What a strong answer sounds like: A strong answer treats a duplicate, a reorder and a partial failure as ordinary inputs rather than exceptions.

How the check measures it

3 questions drawn on this axis alone — 2 medium and 1 hard for a Senior target, because the bar is the level. Hard counts for one and a half times a medium, which is what makes the verdict relative:

Medium rightHard rightVerdict
2 of 21 of 1Ready
1 of 21 of 1Ready
2 of 20 of 1Borderline
0 of 21 of 1Borderline
1 of 20 of 1Gap
0 of 20 of 1Gap

Three questions is coarse and the report says so beside every verdict. It is enough to separate this axis from the other 7, which is what the check is for — there is no total, no percentage and no single number anywhere in the report.

What there is to practise

163 questions on this axis
106 medium, 57 hard · 1 runnable Go exercises. Easy is not drawn for a Senior target.

Every answer cites the source it was checked against.

Drawn from System Design (71), Messaging & Event Streaming (23), Databases & SQL (19), AWS (18), AI Agents (9).

Practised in the System design round
Design a system to requirements and defend the trade-offs: storage, messaging, caching, capacity and cost. The round weighted most in 2026 loops.

Practice is arranged by interview round rather than by topic, because loops are graded that way.

Questions here most often also test system design & trade-offs, data modelling & storage, reliability & failure modes and technical judgment: decisions and trade-offs. Nothing on this axis is asked in isolation, which is why the check reports every competency separately rather than averaging them.

One of them, in full

A real question from the free demo, not written for this page. A demo quiz opens with it; the answer, the explanation and the source come when you submit — no account needed.

Distributed systemsmedium

A nightly batch job splits its work into thousands of chunks across a cluster. One night a single chunk stops making progress while the rest finish, and the job does not complete. The on-call engineer's instinct is to kill the job so it restarts, on the reasoning that the blockage is probably a transient scheduling artefact. What makes that instinct expensive, and what does the situation actually call for?

Answer it on the demo

One axis is not the loop

A Senior backend loop tests 8 competencies, and being strong here says nothing about the other 7. The check covers all of them in one sitting and reports each separately, so what comes back is a profile rather than a score.

See the passesWhat the whole Senior backend loop tests