"Why should we hire you?" usually shows up at the end of the round. Candidates treat it like a softball and give generic answers. The panel treats it like the closing statement. That mismatch is where offers get lost.

The right structure is a value anchor: their specific problem, your specific win, the bridge between the two. 60 to 75 seconds. Pre-written but spoken naturally.

What the interviewer is actually asking

They're asking: "Of the candidates I'm seeing this week, why is the one in front of me the answer to a problem I have right now?" That's a specific question. It rewards specific answers.

Generic answers like "I'm hardworking and I learn fast" don't tell the panel anything they couldn't have guessed. The question is your last chance to surface the one detail that makes you obviously the right hire.

The value-anchor structure

Their problem. Name it. From the job description, the team page, or something the interviewer said earlier in the round.
Your matching win. One specific project where you solved the same shape of problem. With a number.
The bridge. Why the win translates. What you'd reach for on day one.

Worked example (platform engineer)

"From the job description and what you said earlier, the team is trying to cut deploy time from 40 minutes to under 10, and you're rebuilding the CI pipeline to do it."

"I shipped almost exactly that at my last company. We had a 35-minute Bazel build that we got to 7 minutes by parallelising the test sharding and moving the dependency graph build into a warm cache. About half the work was tooling, half was getting twelve teams to migrate their test suites. I owned both halves."

"On day one I'd want to look at where your build spends time, then talk to the teams whose tests are the long pole. That ordering matters because the easy 30% wins are in the tooling, the hard 70% is in the migration. You can't get to under-ten without both."

This answer works because it names their problem (build time, CI rebuild), connects it to a specific personal win (40min → 7min with two specific levers), and ends with the candidate already thinking like a teammate on day one.

Worked example (early-career, ML)

"Your team's hiring an ML engineer to ship experiments faster on the ranking model. That maps to the project I shipped at my last role."

"I built the offline-online consistency check that let us run new ranking experiments in days instead of the prior two-week cycle. Specifically, I wired the offline metric pipeline to mirror the online serving stack, so we caught drift before the model shipped. We went from one experiment per quarter to six per month."

"I'd come in expecting to do the same kind of work here. The first thing I'd want to see is whether your offline metrics correlate with the online outcome you care about. That's where the cycle time hides."

The generic answers that lose

From the candidates we coach, these answers consistently underperform:

"I'm a fast learner." Every candidate says this. It's true of all of them. It signals nothing.
"I work hard and I care about quality." Same problem. Reads as default.
"I match all the requirements in the job post." The panel has already checked the requirements. They're asking the other question.
"I'd be honoured to work here." Skip. The panel is choosing between you and someone else; flattery doesn't move the choice.

The preparation that works

Spend 20 minutes the day before. Three steps:

Read the job posting. Underline the one or two responsibilities they put first. That's their actual problem.
From your last two years of work, pick the project that most closely matches that problem. The match doesn't have to be exact. It has to share the same shape (same constraints, same trade-offs).
Write three sentences. Their problem. Your win with one number. What you'd do day one. Practise it out loud once.

That's the whole answer. It's not flashy. It's the kind of answer that decides ties.

Practise the close in a mock round

LastRound AI mocks behavioural rounds with the closing question as a fixed slot. Get feedback on whether your value-anchor is landing or sliding toward generic.

LastRound AI

"Why Should We Hire You?": Map Your Wins to Their Problem