The 90-Second "Tell Me About Yourself" Answer That Doesn't Sound Like a Resume
"Tell me about yourself" is the most-asked question in tech interviews and the most poorly answered. Candidates either read their resume aloud for four minutes or freeze and mumble for thirty seconds. Both ways lose the room before the real interview starts.
The fix is 90 seconds, structured in three parts. Past, present, why-here. That's it. Practise this once and you'll never wing the opening question again.
What the interviewer is actually asking
They're not asking for your life story. They're asking for context. Specifically: are you a credible candidate, do you communicate clearly under low-pressure prompts, and do you know why you're in this room.
The question is intentionally vague because the panel wants to see what you choose to talk about. That choice is the signal. Spending two minutes on your high-school robotics club tells them you don't know what's relevant. Spending ten seconds on a job you held for three years tells them you can't synthesise.
The 30/30/30 structure
Three buckets, 30 seconds each. Slightly less on the past, slightly more on the why-here if you have to flex.
- 30s, past: One sentence on what you studied or where you started, then one or two sentences on the role(s) that got you to today. Skip the chronology. Pick the through-line.
- 30s, present: Your current role in concrete terms. What you build. What you've shipped lately. One specific number if you have one.
- 30s, why here: One specific reason this company and this team. Not "I love your mission". Reference a real product decision, blog post, or open-source repo of theirs.
A worked example (backend engineer)
Here's what a real 90-second answer sounds like for a mid- level backend engineer interviewing at a payments company:
"I studied computer science at IIIT Hyderabad and started out writing data pipelines for a fintech in Bangalore. I've spent the last three years at Razorpay on the payments-routing team, building the service that picks which payment processor a transaction goes to in real time. The most interesting thing I shipped last quarter was a failover system that cut our checkout drop rate by about 18% during processor outages."
"I'm here because I read your engineering post on multi-region idempotency last month and your architecture is almost identical to the one I've been working on. The scale you're at is the next step up for me, and the routing problems you're solving are exactly what I want to spend the next few years on."
Notice what this answer does. One specific previous role, one specific number, one specific reason for this company. It buys the interviewer thirty real questions they can ask follow-ups on. It also signals that you've read their engineering blog, which most candidates don't.
What candidates get wrong
Across the candidates LastRound AI sees in mock rounds, four failure modes account for most weak openers:
- The chronology dump. "I graduated in 2018. Then I joined company X. Then I moved to company Y in 2020. Then..." Nobody's tracking your timeline. Pick the through-line.
- The mission-statement opener. "I'm passionate about technology." Skip this. Everyone says this. It signals nothing.
- The numbers vacuum. Saying "I worked on a recommendations system" is half an answer. Saying "I worked on a recommendations system that served 4 million users at 30 QPS" tells the panel you can talk about your work concretely.
- The generic why-here. "I want to work at a company that values engineering excellence." This fits any company on Earth. Replace it with one specific thing about THIS company.
How to prepare the why-here
The why-here sentence is the highest-payoff one in the whole answer because nobody else will have it. Spend fifteen minutes the day before doing this:
- Read the company's most recent engineering blog post. Note one specific technical decision they made.
- Skim their LinkedIn job posting for the role. Find the one responsibility that overlaps most with what you've shipped.
- Bridge the two in one sentence. "Your post on X connects to a problem I solved at Y."
That sentence ends your answer. The panel will almost always follow up on it, which means you've already steered the conversation toward your strongest topic.
Practise out loud, not in your head
The biggest mistake people make is rehearsing this question silently. It sounds fine in your head and falls apart out loud. Time yourself. 90 seconds feels longer than you think when you're talking, shorter than you think when you're listening. Record one take and play it back. You'll cut twenty percent of words on the second pass.
Written by
Shekhar
LastRound AI
On the LastRound AI team. Writes about career advice, behavioral interviews, and how to navigate hiring at startups and big tech.
Further reading
- The Muse — STAR method guide — How to structure behavioural answers
- Tech Interview Handbook — Free, open-source interview prep
- Cracking the Coding Interview — Industry-standard prep book reference
Share this post
Related articles
Interview Q&A
50 API Developer Interview Questions & Answers 2026 (REST, GraphQL, Design) | LastRound AI
Interview Q&A
35+ Account Executive Interview Questions 2026 (Enterprise Sales) | LastRound AI
Interview Q&A
30+ Blockchain Developer Interview Questions & Answers 2026 (Smart Contracts & Web3) | LastRound AI
Interview Q&A
35+ Business Analyst Interview Questions That Actually Matter (2026 Guide) | LastRound AI
