Three questions show up in almost every interview, yet get the laziest "I'll wing it" prep:
- "Why do you want to join us?"
- "Why do you want to work in AI / ML?"
- (Closing) "Do you have any questions for us?"
The first two test motivation and fit — answer vaguely and the interviewer thinks "you're mass-applying"; answer concretely and you signal "I did my homework and I'll stick around." The third is your last bit of agency: the quality of your questions directly shapes the interviewer's final impression.
This post gives you a framework, templates, and landmines for all three, plus a ready-to-use list of questions to ask.
① "Why do you want to join us?"
The interviewer wants to confirm: are you genuinely interested in this company, or just need a job? Your answer should prove you did your homework and tie "the company's specifics" to "your motivation."
Three-part framework:
A concrete thing about the company (product/tech/culture) → why it resonates with me → what I can bring
Landmines:
- ❌ Generic filler: "you're an industry leader with great benefits" — true of any company = no homework.
- ❌ Only what you'll get (learning, salary), nothing about what you'll contribute.
- ❌ Getting facts wrong (misstating product lines/recent news) — worse than not mentioning them.
Template (English):
"I've been following your work on
[specific product/tech direction], especially[a concrete thing — a feature, an open-source project, an eng blog post]. What draws me is[tie to your interest/values]. My background in[relevant experience]should let me contribute quickly on[a team/direction]."
💡 Prep move: spend 20 minutes before the interview reading their engineering blog, recent product launches, and between the lines of the JD, and find 1-2 concrete points that "only make sense for this company." That's the whole key.
② "Why do you want to work in AI / ML?"
The interviewer wants to confirm: are you chasing the hype, or do you have sustained passion and understanding? The most convincing answer is a real turning point + evidence of ongoing investment.
Three-part framework:
A trigger (what got you hooked) → what you kept doing (proof it's not a fad) → what problem you want to solve next
Landmines:
- ❌ "Because AI is the trend / pays well" — honest but mercenary and replaceable.
- ❌ Empty passion with zero evidence (no projects, no continued learning).
- ❌ Overhyping AI as a cure-all, which reads as naive.
Template (English):
"What really got me in was
[concrete trigger — a small project, an application I saw], when I first felt[what problem AI solves that excites you]. Since then I've kept[concrete investment: projects built, things learned, what I read]. The direction I most want to go deep on is[specific area — LLM serving / recommenders / CV], because[why this problem matters to you]."
💡 Key: replace adjectives with evidence. "I'm passionate about ML" convinces no one; "I built X in my off-hours, hit pitfall Y, learned Z" — that's passion.
③ "Do you have any questions for us?" — The Reverse Question Is a Scoring Question
Never say "no." Saying no = you don't care that much. Good questions show your depth of thinking, seriousness about the role, and that you're evaluating them too (interviews are two-way).
Three principles for your questions:
- Ask what you can't Google — things only an insider knows, not what's on the website.
- Tailor to the audience — ask engineers about tech/daily life, managers about team/growth, HR about process/culture.
- Show you're already thinking "once I'm in" — let them picture you as a teammate.
A Ready-to-Use List of Questions
🔧 For engineers / future teammates
- "What's the team's biggest technical challenge right now?"
- "What does a feature's path from idea to production look like?"
- "What's the culture around code review / testing / deployment?"
- "What's one thing you love and one thing you'd change about working here?"
👔 For the manager / hiring manager
- "What does doing well in this role look like in the first three to six months?"
- "What's the team's most important goal for the next year?"
- "How do you think about growth / promotion paths for people on the team?"
- "What capability is the team most missing right now?" (hear the pain they're really solving)
🤖 For AI/ML roles (bonus)
- "What's the path from research to production for a model? How do research and engineering collaborate?"
- "How do you measure an ML project's success? How do you balance offline metrics with online business metrics?"
- "How mature is the data / compute infrastructure today?"
🏢 For culture / HR
- "How does the team make technical decisions — top-down or engineer-driven?"
- "What's the real situation with remote / hours / on-call?"
- "What do the next interview steps and timeline look like?" (a practical, graceful close)
💡 Closing technique: prepare 5-6 questions (some may get answered during the interview, so have backups). After asking, add a sincere wrap-up: "Thanks — after this conversation I'm even more interested in the role, especially
[a specific point]." That maxes out the final impression.
The Shared Core of All Three
| Question | What it really tests | Winning key |
|---|---|---|
| Why this company | did you do homework, how long you'll stay | find a point "only this company" fits |
| Why AI/ML | real passion or hype | replace adjectives with evidence (projects/investment) |
| Any questions | depth, two-way evaluation | ask the un-Googleable, show "once I'm in" |
What they share: concrete > vague, evidence > adjectives, two-way > one-way flattery. Interviewers hear dozens of canned answers a day — just being sincere + having done your homework + naming concrete specifics already beats most candidates.
Wrap-Up
"Why us," "why AI/ML," and "any questions" look like pleasantries but are high-weight signals of your fit and sincerity — and they're entirely preparable. Spend an hour per company customizing the "concrete points" for the first two, and walk in with your own list of questions, and you turn the three most-neglected questions into your scoring positions.
This is part of the interview series, and together with the earlier 2-minute self-introduction framework and the STAR behavioral series (overcoming challenges, fast learning, ownership, failure lessons, ambiguous requirements), it forms a complete interview prep kit.