© Interview RuntimeThis content is copyright. Copying is disabled; you may quote a short excerpt with a link to the page.

Behavioural interview practice for mobile engineers, with 6 stories to say out loud

6 decisions from your own experience, rehearsed with interruptions that test your contribution, then retold until the story holds under questions.

Engraving of 6 small wooden drawers in a cabinet, one pulled open.

In short

  • Choose stories where a reasonable engineer could have decided differently, and name the call you made.
  • Describe the other person's position in a form they would recognise before you describe the outcome.
  • A detail you cannot recall is a reason to narrow the claim, not to fill it in.
  • Each story needs evidence from before the decision, not only the result.
  • Retell each story after an interruption, and check whether the decision still holds.

“Who could stop the rollout?”

It is an awkward question if your release story begins with “we decided” and ends with “we shipped”. There may be 4 people inside that we. One saw the regression. Another understood which customers were affected. Someone owned the release switch. Someone had to explain the delay. Which of those jobs was yours?

Behavioural interview practice for mobile engineers starts with 6 decisions you can reconstruct from experience in a senior iOS or Android round. Several can come from the same project. Choose episodes where a reasonable engineer could have made a different call. Those are easier to discuss when an interviewer interrupts the chronology and asks what you knew on Tuesday afternoon.

The scenes and sample answers in this article are hypothetical. The follow-up is where these stories are tested, as the follow-up decides the senior mobile interview explains.

Use each story in the same loop. Answer the question in 2 minutes, take the interruption, note where you became uncertain, then retry after the detail changes.

1. Hard decision, a release you slowed down or narrowed

Question to rehearse. Tell me about a time you traded delivery speed for reliability. What information justified the cost?

Evidence it needs. The release decision point, what was already on users' phones, what the team could change remotely, and who had to wait for a store update.

Pick a release with an identifiable decision point. What was already on users' phones? What could the team change remotely? Who had to wait for a store update? A feature flag might hide a broken screen while leaving a database migration in place. If the distinction affected your decision, it belongs early in the story.

Picture a release where a team is preparing to switch its downloads screen to a new local index. Fresh installations pass. An upgrade test leaves previously downloaded files on disk but removes them from the screen. The mobile engineer asks to hold the migration for existing installs and shows the release owner a device on which a reader has to download a book again. The owner accepts a narrower release. That account gives a listener enough to ask whether the team could reliably distinguish upgrades, and what happened to anyone already migrated.

Follow-up. What would have made you comfortable shipping the original plan? Prepare the evidence you asked for, who supplied it and the boundary of your authority.

Retry. Tell the story again in 2 minutes, then answer the follow-up before rereading the decision point.

If a follow-up exposes a gap in how you explain a decision, Guided Practice in iOS Interview Training or Android Interview Training lets you revise the answer after a changed constraint.

2. Technical conflict, a disagreement where the other person had a good argument

Question to rehearse. Describe a technical disagreement with someone whose judgement you respect.

Evidence it needs. The other engineer's strongest argument, the moment the discussion moved, and who made the final decision.

A common weak spot is presenting the alternative so badly that the choice sounds obvious. Give the other engineer a position they would recognise. Suppose a team wants a shared navigation contract and another team wants platform-specific flows. A shared contract could keep analytics events consistent across clients. Separate flows could preserve existing back behaviour and release independence. Both positions have engineering merit.

Find the moment that moved the discussion forward. Perhaps you each implemented 1 deep link into an interrupted checkout. Perhaps product clarified that a campaign link had to reopen an existing cart. Explain what you learned, what you conceded and who made the final decision. “We aligned” leaves out the difficult part.

Follow-up. Where did the rejected approach remain better? An honest answer might name 1 maintenance cost your team accepted. If the disagreement is still unresolved, describe the interim rule and who can revisit it.

Retry. Tell it again, this time stating the rejected approach's strongest point first.

3. Incident, a production problem where your first explanation was wrong

Question to rehearse. When did evidence make you change your mind?

Evidence it needs. The observation that eliminated your leading explanation, the timeline in order, and what was still unknown during mitigation.

Choose a case where you can distinguish a symptom from a cause. Say users report an empty inbox after an update. That could mean local messages were deleted, the query used the wrong account, or the screen replaced cached content after a failed refresh. Each possibility changes what you do for customers. State the observation that eliminated your leading explanation.

Rebuild the timeline in the order it happened. At detection, what could you see? During mitigation, what was still unknown? After recovery, what did the team learn? Compressing all 3 into a polished explanation makes it sound as if you had hindsight at the start.

Follow-up. What did you do before you understood the cause? If support handled customer communication while you isolated affected versions, give support the credit. If you lost time on the wrong hypothesis, say which signal you would collect earlier next time.

Retry. Tell the timeline in the order it happened, without the final cause, and let the interviewer ask what you knew at each step.

4. Influence, a requirement you changed after seeing the user consequence

Question to rehearse. Tell me about a time you influenced what the team built.

Evidence it needs. The failure you recorded or reproduced, the acceptance criterion or delivery sequence you changed, and the cost you accepted.

Start with the requested behaviour in ordinary language. Take “Mark a message as sent immediately”, which sounds reasonable until the phone goes offline between queuing the message and receiving a server acknowledgement. The question for product is what the sender should believe at that moment. A short recording of the failure can change that conversation more quickly than a diagram.

Use a real example where your intervention changed an acceptance criterion, the sequence of delivery or the decision to proceed. Then describe the cost. A pending state needs copy, accessibility behaviour, retry rules and support guidance. If another person raised those issues, make that visible.

Follow-up. What would have happened if product had declined your proposal? Describe the version you could still have shipped responsibly. Avoid inventing customer losses or attributing revenue to a change you never measured.

Retry. Present the failure recording first, then the proposal, and check whether the cost is still stated.

5. Delivery under pressure, a cross-team change that had to work across app versions

Question to rehearse. When did you influence delivery outside your immediate team, with a fixed release window?

Evidence it needs. The window that forced the decision, the older app version that still needed support, and the check that showed it was safe to remove the compatibility code.

A backend can deploy tonight while an older mobile binary stays in use. That asymmetry can provide a more revealing story than the number of people in your architecture meeting. Which old behaviour still needed support? Which team owned the transition? What let them tell that it was safe to remove compatibility code?

Imagine a proposed change that introduces a new status for an order. An older client may display an unknown value badly. Persuading the service team to delay the new value is only the first decision. The stronger account explains how the teams agreed on fallback behaviour, exercised an older build and assigned ownership of eventual removal. Use that level of detail only where it matches your own work.

Follow-up. Who changed their plan because of your contribution? A precise answer can be modest. “I wrote the compatibility test that the API team added to its release check” tells a listener more than “I drove cross-functional strategy”.

Retry. Tell it again with the release window stated first, then describe which older version you tested.

6. Ownership, an improvement someone else could own after you left

Question to rehearse. What became easier for another engineer because of something you did?

Evidence it needs. What another engineer did without you, when it happened, and what still depended on you.

Look for a transfer of judgement. Perhaps 1 person used to diagnose every flaky UI test. Pairing through a failure, documenting how to distinguish a bad fixture from a genuine timing issue, and watching another engineer diagnose the next failure are 3 different contributions. State which ones happened.

Choose an outcome you could substantiate. A later pull request from someone else, an incident handled independently or a checklist used on the next release can all be meaningful. You can describe what you observed without exposing confidential material. If all you know is that you wrote a guide, stop there. Its existence does not establish adoption.

Follow-up. What still depended on you? That boundary makes the claimed outcome easier to trust and gives you something specific to improve.

Retry. Tell it again starting with what the other engineer did next without you, then go back to what you did before it.

Turn 1 story into a 2-minute answer

On paper, write 5 short sentences covering what the team needed, what you owned, what you knew at the time, what you chose or recommended and what you later observed. Then cross out the sentences and tell the story aloud. Spend the most time on the decision the interviewer can question.

This article does not teach a fixed answer structure. If you want to compare with the STAR format, the STAR method for senior mobile engineer interviews explains it.

Here is a weak first attempt. “I led a migration, worked with stakeholders and improved reliability.” Try interrupting it at any word. Even led needs an explanation.

A more inspectable answer. “I owned the iOS upgrade path. On a device upgraded from our previous release, downloads disappeared from the library even though the files were still present. I showed that case to the release owner and recommended holding the index change for existing installs. She approved the hold. I added the upgrade fixture. Another engineer fixed the index rebuild. The fixture passed on that device before we resumed. I don't have a customer-wide recovery figure.”

Now interrupt after the recommendation. Why a hold? Which other versions were tested? Who could resume the release? Those questions have a place to land. In your own answer, a detail you cannot recall is a reason to narrow the claim.

Rehearse the interruption

Give a partner a single instruction. Interrupt after the first “we” and ask who did what. On the second run, have them ask what evidence existed before the decision. On the third, ask what another team member would describe differently. If all 3 answers are identical, you may be repeating the project summary.

Before the interview, keep 1 line per story with the decision, your contribution and 1 observed consequence. Leave out client names, private dashboards and sensitive incident details. A useful note should help you find the memory without scripting the answer.

For the interviewer's perspective, read what goes into a senior mobile interview debrief.

Continue with structured practice

iOS Interview Training and Android Interview Training are $99 each, paid once per platform. The Mobile Bundle covers both for $169. Each book and its companion code are yours to keep, and each includes 90 days of Guided Practice. You answer realistic prompts, inspect weaknesses, revise your answer, handle changed constraints and compare attempts. It is not a live mock interview with a person. Read the free iOS sample or Android sample before you buy, and start with the platform you will interview in.

© 2026 Mike Salari, Interview Runtime. All rights reserved. Quoting a short excerpt with a link to this page is welcome; republishing the article, in full or in part, is not.

Practice the follow-up, not the first answer.

Interview Runtime is the interview book plus 90 days of Guided Practice: you answer first, see the one gap that would cost the point, then answer again when one fact changes.