Questions to Ask the Interviewer

Save the questions only an owner of the business can answer: funding stage and months of runway, how the company will make money, what stops a competitor copying it, and which people the business depends on.

Asking a hiring manager to define success at 90 days converts an advertised job description into the standard you will actually be judged against, and it reveals whether the manager has shaped the role at all.

Ask a skip-level director the org-level questions only they can answer: where this team's work sits among company priorities, how headcount is planned, how disputes between teams get decided. Day-to-day questions belong in the peer round.

Save the process and calibration questions for the recruiter: how many stages there are, what each assesses, the timeline, the level the role is posted at, and the pay band for that level. Later interviewers rarely own those facts.

A per-round bank matches each question to whoever can actually answer it: a recruiter knows process and timeline, an engineer knows daily practice, a manager knows scope. One generic list burns a five-minute window on the wrong person.

Asking nothing reads as low interest, not efficiency. A peer engineer hears that the candidate is not evaluating the team back, and the write-up often records it as weak engagement, since every other candidate that day used the same window.

A question the careers page already answers spends your closing minutes proving you did not prepare. Panels read it as low interest and weak filtering, and it burns your one chance to ask something only an insider knows.

A hedge answers in the abstract and never lands on a specific, recent, checkable instance. Listen for generalities, drift toward what the company intends to do, and answers pitched at the whole company when you asked about this team.

A candidate's closing questions are evidence, exactly like their answers: they show what that person has had to care about at work. Questions about equipment and perks describe a scope nobody has had to own yet.

Ask plainly and specifically: what stage the last raise was, how many months it funds at the current plan, and what milestone unlocks the next round. Framing it as planning your own bet keeps the question neutral.

Bound the question. Name the team, fix a time frame, force a ranking and ask who sets it: 'Where does this team's work sit in next year's top three priorities?' Open-ended strategy questions return the public mission instead.

Ask for a range attached to a named level, not a single figure: "What is the band for this level, and how is level actually decided?" That yields the range plus the mechanism that decides your rung.

Rank by two things: what only this interviewer can answer, and what would most change your decision. Ask the top live question first, leave room for the answer and one follow-up, and let the rest go rather than racing.

Ask for the most recent instance instead of the general policy. A follow-up that requests one episode — what happened, when, and who decided — converts an abstraction into something checkable, and one follow-up per hedge is the right budget.

It signals the candidate has lived with a decision being reversed. The subject is the team's process over time rather than the asker's own comfort, and the revisiting clause is the part that cannot be answered with a slogan.

The walkthrough from merge to user exposes hand-offs, approvals and who is allowed to deploy. It turns a vague sense of engineering culture into a countable path, and the number of waiting steps predicts what the job feels like.

Ask for a count and a recent instance instead of an opinion: how many share the rotation, how many deploys went out last week, how long the last change waited for a first comment. Recent countable facts resist rehearsed reassurance.

Ask who pays today, what they pay for, and who is expected to pay next; then ask what would stop a better-funded competitor from copying the product within a few months. Specific answers beat vision.

Ask for mechanics rather than adjectives: the cadence of your one-to-ones, what gets written down and when, how the last person on the team made a promotion case, and how you would first hear that something was off track.

Answers fall into three shapes — funded new work, a backfill after someone left, and a seat re-opened after a search or hire did not work out — and each one calls for a different second question. The hedging around the answer often carries more signal than the answer itself.

Ask for a funded example, not a value: which non-feature engineering work was resourced in the last planning cycle, who worked on it, and what it displaced. Stated commitment to quality costs nothing; protected capacity is the evidence.

Anything requiring firsthand engineering or team experience: how systems are built, how tradeoffs were made, review and on-call culture, the manager's real style, why a specific person left. Recruiters own the requisition and the process, not the day-to-day work.

Strike it from the card and move down to the next prepared question, or deepen what was already said into a specific follow-up. Name the adaptation out loud rather than asking a question that was answered twenty minutes ago.

In round one nobody has decided they want you yet, and a technical interviewer usually cannot answer pay or leave questions anyway. Early it reads as package-first; later, with whoever runs the hiring process, it is routine.

Treat it as one signal, not a verdict. Note what was offered instead — total raised, investor confidence, a growth story — ask once more for months, then corroborate against hiring pace, recent departures and how the offer is pitched before drawing a conclusion.

A usable answer names a decision owner or forum, gives a recent real dispute, states what was decided and says what the losing team dropped. Warmth without a mechanism means contested work goes to whoever pushes hardest.

Treat it as a phrase that needs a follow-up, not a nod. It describes a feeling rather than a mechanism, so ask what it cashes out to in practice: how disagreement gets settled and what happens when work runs late.

Anchor the ask to countable recent incidents and to artefacts: how many of the last several incidents produced a written postmortem, who wrote them, and what changed afterwards. A question about whether the culture is blameless invites a policy answer instead.

Step down one rung, not sideways: ask how long their last change waited for a first comment, then whether that week was typical. Two follow-ups, then move on and ask another interviewer the same question rather than pressing.