Start with the domain contract. Specify tenant, project, canonical URL, crawl configuration and run ID before deduplicating or comparing results. A strong answer names stable identities, version rules and the evidence that confirms a durable outcome before selecting infrastructure.
The official sources describe semrush provides online visibility, SEO, content and competitive-research tools built from web and marketing data. They support product context, not a fixed interview loop or the exact questions in this guide.
Reason about state transitions. Keep crawl runs immutable and publish a new report pointer conditionally so a late run cannot replace a newer result. Put tenant or account boundaries into keys and queries rather than relying on a caller to remember them.
Design the recovery path with the happy path. Record input scope, rule version and watermark so a metric can be recalculated and explained after data arrives late. State what can be retried, what must be looked up first and what requires human review.
Explore your preparation priorities
Choose a focus to see how to prepare.
Define a crawl identity
Specify tenant, project, canonical URL, crawl configuration and run ID before deduplicating or comparing results.
YOUR PREPARATION- Name the stable identity and tenant boundary.
- Write the success and replay invariants.
PracHub practice map for Semrush. Select a checkpoint to connect the domain contract, state boundary and recovery decision to a practice prompt.
Define a crawl identity
editorialSpecify tenant, project, canonical URL, crawl configuration and run ID before deduplicating or comparing results.
What to demonstrate
- Web-data correctness
- Clear trade-off reasoning
How to prepare
- Draw the state and identity boundaries.
- Add a replay, conflict or lost-response case.
Protect report versions
editorialKeep crawl runs immutable and publish a new report pointer conditionally so a late run cannot replace a newer result.
What to demonstrate
- Crawl scheduling
- Clear trade-off reasoning
How to prepare
- Draw the state and identity boundaries.
- Add a replay, conflict or lost-response case.
Make analytics reproducible
editorialRecord input scope, rule version and watermark so a metric can be recalculated and explained after data arrives late.
What to demonstrate
- Reproducible analytics
- Clear trade-off reasoning
How to prepare
- Draw the state and identity boundaries.
- Add a replay, conflict or lost-response case.
PracHub editorial advice for the preparation topics above.
Follow the state, not just the happy path
Choose a scenario to trace what changes.
The expected version still matches the stored state.
- 01Edit version 3Edit version 3.
- 02Compare versionCompare version.
- 03Save version 4Save version 4.
Commit one new version and return durable confirmation.
Compare three outcomes for Semrush practice: a confirmed write, a version conflict and a lost response.
Choosing infrastructure before defining identity and state
Name tenant scope, stable IDs, versions and durable confirmation before selecting queues, caches or databases.
Treating a timeout as proof of failure
Model the result as unknown, look up the original operation and reuse the same identity.
Joining multiple one-to-many tables at detail grain
Aggregate each input to the requested output grain and assert row-count invariants.
Quoting an unverified interview sequence
Use the exact recruiter instructions and role posting; describe this guide as preparation rather than employer process evidence.
Choose a category, try a prompt, then open its approach, worked solution or follow-up when you need it.
Merge crawler activity windows
Given half-open crawler activity windows [start, end), merge overlapping or touching windows. Reject an end before its start and return a sorted new list.
Approach
- Sort by start, then scan while keeping one current interval.
- Merge when the next start is at or before the current end; validate every interval before returning a fresh result.
Worked solution 35 min
- Sort the intervals.
- Validate each boundary during the scan.
- Merge overlap and exact touching; otherwise start a new result interval.
def merge_windows(windows):
merged = []
for start, end in sorted(windows):
if end < start:
raise ValueError("end before start")
if not merged or start > merged[-1][1]:
merged.append([start, end])
else:
merged[-1][1] = max(merged[-1][1], end)
return [tuple(item) for item in merged]
Scroll sideways to view long lines.
Follow-up
- How would an inclusive end change the touching-window rule?
Deduplicate a tenant-scoped crawl frontier
Accept tenant, project, normalized URL and depth records. Preserve the shallowest depth, reject cross-tenant key collisions and return URLs in stable first-seen order.
Approach
- Key deduplication by tenant and event ID, then store a fingerprint of accepted content.
- Keep ordering/version rules separate from duplicate detection and make acceptance atomic.
Follow-up
- How long must replay evidence be retained, and what happens after expiry?
Count recent crawl failures
Implement add(timestamp) and count(now) for failures in (now - 10, now]. Timestamps are nondecreasing and equal timestamps represent separate failures.
Which events still count?
The left boundary is excluded; the right boundary is included. Drag past an event to see it enter, then expire 10 seconds later.
See the event values
- At 0s: +1 — expired
- At 5s: +1 — in window
- At 10s: +1 — in window
- At 14s: +1 — not arrived
- At 19s: +1 — not arrived
Synthetic crawler activity windows events at 0, 5, 10, 14 and 19 seconds. Drag the clock: the interval is (now - 10, now], so the lower boundary is excluded.
Approach
- Keep timestamps in a deque and remove values at or before now - 10 before returning the count.
- Reject a backward clock; every timestamp enters and leaves once, so updates are amortized O(1).
Follow-up
- How would you bound memory for many inactive keys?
Find the latest crawl result per URL
Given crawl_results(tenant, url, run_id, finished_at, status), return the latest result per URL for tenant a. Break timestamp ties with run_id.
Approach
- Filter to the tenant first, then rank within the full tenant-scoped business key.
- Order by event time and a deterministic unique tiebreaker, then select rank one.
Worked solution 35 min
- Filter to tenant a before ranking.
- Partition by tenant plus entity ID.
- Order descending by time and event ID, then keep row one.
WITH ranked AS (
SELECT tenant, url, status,
ROW_NUMBER() OVER (
PARTITION BY tenant, url
ORDER BY finished_at DESC, run_id DESC
) AS rn
FROM crawl_results WHERE tenant = 'a'
)
SELECT url, status FROM ranked WHERE rn = 1 ORDER BY url;
Scroll sideways to view long lines.
Follow-up
- When should arrival time replace event time for operational views?
Aggregate issue counts without multiplying page metrics
Join pages, crawl issues and keyword observations. Return one row per page with issue and keyword counts without a many-to-many multiplication.
Approach
- Aggregate each one-to-many input to its target grain before joining.
- Join on the full tenant-scoped key and use a left join where missing activity must remain visible.
Follow-up
- Which invariants would detect silent row multiplication?
Design a polite multi-tenant crawl scheduler
Design URL discovery and fetching with robots rules, host budgets, retries, fairness, deduplication and immutable run evidence.
Approach
- Name the stable operation identity and authoritative state owner before drawing services.
- Persist state changes and outbound work together, make consumers replay-safe and expose an explicit status read.
Worked solution 35 min
- Write the operation and state invariants.
- Trace success, exact replay, conflict and lost response.
- Add authorization, observability and a reconciliation queue.
Follow-up
- Which failure needs reconciliation rather than automatic retry?
Design a versioned visibility-report pipeline
Combine periodic keyword observations and crawl results into reproducible reports while handling late data and corrected rules.
Approach
- Partition by the entity whose order matters and attach tenant scope to every key.
- Define watermarks, retry budgets, backpressure and a replay path before choosing throughput numbers.
Follow-up
- How would you rebuild a derived view without repeating external side effects?
Stop an older crawl from replacing a newer report
Run A is slow, run B completes and publishes, then A overwrites the project pointer. Reproduce the race and specify the conditional publication repair.
Approach
- Capture both operation IDs and versions in a deterministic test that resolves responses in reverse order.
- Guard the authoritative write atomically and verify identity/version again when the UI applies a result.
Worked solution 35 min
- Build a test that controls response order.
- Log operation identity, expected version and resulting version.
- Add atomic server guards and stale-result suppression at the caller.
Follow-up
- What telemetry distinguishes a harmless stale response from state corruption?
A seven-session PracHub practice plan with one reviewable artifact per session. Adjust the pace to your experience and interview date; it is not a company hiring timeline.
Prepare, practise & reflect
One practical outcome each day. Spend longer where you need it.
0 / 7 done01Choose the role boundary
- Read the exact opening.
- List the product, service and operational responsibilities it names.
Deliverable: A one-page role-scope note
02Practice boundary-aware code
- Run the interval solution.
- Add touching, contained, empty and invalid cases.
Deliverable: A tested boundary contract
Practice prompt ↗Worked solution ↗03Protect event identity
- Model exact replay and conflicting reuse.
- State the atomic write boundary.
Deliverable: An idempotency table and invariant list
Practice prompt ↗Practice prompt ↗04Verify SQL grain
- Run the latest-state query.
- Draw the row grain before the aggregate join.
Deliverable: SQL results plus cardinality notes
Practice prompt ↗Practice prompt ↗Worked solution ↗05Design recovery first
- Trace success, conflict and lost response.
- Name the reconciliation owner.
Deliverable: A failure-path sequence diagram
Practice prompt ↗Practice prompt ↗Worked solution ↗06Reproduce stale state
- Resolve two requests in reverse order.
- Add server and client generation checks.
Deliverable: A deterministic regression test
Practice prompt ↗Worked solution ↗07Rehearse evidence-based stories
- Practice two real examples aloud.
- Remove team-level claims you cannot attribute to your action.
Deliverable: Two concise STAR notes with measurable evidence
Practice prompt ↗Practice prompt ↗Expand any day for tasks and deliverables. Your progress is saved on this device.
Use real examples. Name your responsibility, the evidence available at the time, the trade-off and what changed after your decision.
Explain a misleading metric you corrected
Describe how you found that a plausible dashboard number was wrong or incomplete.
Approach
- Use a real example and name your individual responsibility.
- Explain the evidence, trade-off, action and measurable or observable result.
Follow-up
- What would you do differently with the information you have now?
Resolve a shared-capacity dispute
Describe how you balanced one customer workload against platform health and other tenants.
Approach
- Use a real example and name your individual responsibility.
- Explain the evidence, trade-off, action and measurable or observable result.
Follow-up
- What would you do differently with the information you have now?
Communicate a data-quality incident
Tell a story where results were available but not trustworthy.
Approach
- Use a real example and name your individual responsibility.
- Explain the evidence, trade-off, action and measurable or observable result.
Follow-up
- What would you do differently with the information you have now?
- 01
Bring one result you improved and one decision you changed after seeing evidence.
Are these verified Semrush interview questions?
No. They are PracHub editorial exercises informed by official Semrush product and careers context. Team requirements and interview formats vary.
Semrush — Company ↗Semrush — Developer API ↗Which language should I use for the coding practice?
Use a language accepted for your interview and explain its collection, numeric and error-handling behavior. The Python examples here emphasize the contract rather than a required company stack.
Is this Semrush interview schedule?
No. It is a seven-session PracHub preparation checklist. Follow the timing and format in your invitation.
Sources & methodology 5 sources ↗
Official role evidence, timestamped platform data and clearly labeled preparation advice.
- 01Semrush — Company ↗
Official company and product context; no universal interview sequence is stated.
official · Accessed 2026-09-20 - 02Semrush — Developer API ↗
Official API context for data access and integration; exercises are editorial.
official · Accessed 2026-09-20 - 03Python — urllib.parse ↗
Technical reference for the URL-normalization exercise.
official · Accessed 2026-09-20 - 04PostgreSQL — Window functions ↗
Technical reference for ranking and latest-state SQL. Runnable fixtures use SQLite.
official · Accessed 2026-09-20 - 05PracHub — Software Engineer practice ↗
Cross-company practice; not evidence of company interview questions.
platform · Accessed 2026-09-20
Explore related engineering environments and compare the preparation focus.
Verizon Software Engineer Interview Questions & Guide 2026
high-volume provisioning and observable network state
Thales Software Engineer Interview Questions & Guide 2026
high-assurance embedded and secure systems
TE Connectivity Software Engineer Interview Questions & Guide 2026
hardware-software integration and production traceability
Swedish Space Software Engineer Interview Questions & Guide 2026
telemetry ingestion, pass scheduling and mission reliability