Edit wiki/mentorship/asking-good-questions.md — Work Reflections

---
visibility: public-edit
---

# how to extract value from mentor conversations

over a summer at a [[neurotech startup|signal-processing-workflow]], i had probably 30+ structured conversations with mentors, cofounders, peers, and visitors. some were incredible. some were mediocre. the difference was almost always in the questions.

## what worked

### ask about their experience, not their advice

instead of "what should i do?", the better question is always about what they actually experienced. "how has your experience of running [the company] been? tiring? easy to maintain motivation?" gets real answers. "what should i do with my life" gets platitudes.

### probe the disagreements

the most valuable moments came when i found contradictions between what different mentors told me:
- one cofounder said he cared deeply about intern growth; another called that altruism
- one person said a former colleague was a mistake; the other said they missed them
- everyone had different opinions on each other

"be very careful of what other people tell you about" — this was itself great advice. mentors have biases. the skill is triangulating across multiple perspectives.

### ask the meta question

"what do you think of me?" sounds needy but it's incredibly high-signal. one CEO answered: "cool, fun person. don't know what you want." that single sentence told me more about my blind spot than weeks of self-reflection.

"what can be better about [the org]?" is the same move in reverse — showing the mentor you're thinking critically, not just absorbing.

### use walks

"one of the biggest ways i can get feedback is walks." walks strip away the formality. no screen to hide behind, no agenda pressure. the best conversations happened walking around the city at night or during breaks between work sessions.

## what didn't work

### asking for validation

"you're not just lying to me to not make me feel sad, right?" — i asked this early on and it reveals insecurity more than it generates useful info. the answer is always going to be "no" regardless.

### pre-filtering your questions

i had a habit of crossing out questions before meetings because they seemed too basic or already answered. bad move. the "dumb" questions often led to the most surprising answers.

### not pushing back

"push back more. you're too soft with people you think are better than you." i had a filter coefficient problem — i was setting it too low (accepting everything) for people with more experience, when actually those people have a higher chance of being tunneled in their own perspective.

the fix: "do you understand it well enough to explain it back to your sibling? if not, ask more questions."

### arguing to be right instead of for the idea

"many times it feels that you are arguing to be right instead of for the benefit of pursuing the idea." — this feedback hit hard because it was true. the goal of a question should be to advance understanding, not to win.

## the agenda principle

"whenever you talk to anyone, state one sentence of what you are doing and why." basically: have an agenda. not in a manipulative way — in a clarity way. know what you want from the conversation before it starts.

but also: be open to the conversation going somewhere unexpected. the best insight i got all summer (the [[agency talk|agency-talk]]) came from a question about how my day went, not from my prepared list.

## recording vs not recording

i went back and forth on this. the CEO's take was decisive:

"don't record. there is a single moment to learn, and it is now. it increases focus."

my counter was that recording makes me relaxed and lets me converse better. i ended up taking notes right after instead — trying to capture the emotional texture, not just the content.

---

*see also: [[pattern recognition|pattern-recognition]], [[disagreeing productively|disagreeing-productively]], [[social wins|social-wins]]*