Quant Researcher

Stage by stage

Close each stage out loud, so they always know where you are.

1. 1The exact target5 minThat's the target and when we forecast it. Next, the inputs we'd really have.
2. 2Inputs known at the time6 minThose are the inputs we'd have had at that moment. Now the baseline to beat.
3. 3A baseline5 minThat's the simple answer my model has to beat. Next, how I'd test it honestly.
4. 4An honest test8 minThat's the test, in time order. Last, what would make me drop it.
5. 5What would make you drop it6 minThat's my rule for dropping it, set before I look. Where should I go further?

How people lose this round

• I'd train a deep network on the whole history.It skips the baseline, so you can't tell whether it learned anything. And daily data gives a few thousand rows, little for a model with many weights.
• I'd check it with 5-fold cross-validation.Shuffled folds train on days after the ones they test, and neighbouring days look alike. Test in time order.
• It fits with an R² of 0.6, so it works.A fit on the data it was built from comes cheap. What counts is how far it beats the baseline on data it never saw.
• The best of my 40 versions has a t-statistic of 2.5.Try enough versions and one clears 2.5 by luck. Say how many you tried, and keep data back for the winner.
• My strongest input is the volume over the next few minutes.It's strong because it's nearly the answer, and you can't know it at the moment you forecast.

Stage by stage

Close each stage out loud, so they always know where you are.

1. 1Restate1 minSo we want exactly this. Have I got that right?
2. 2Try a small case3 minThat's the small case. It points to a method for the full one.
3. 3Choose the method2 minI'll set it up this way. Here goes.
4. 4Compute5 minThat's my number. Let me check it before I commit.
5. 5Sanity check2 minIt passes the checks, so that's my answer.
6. 6Take the hint and the follow-up2 minThat helps. Here's how it changes my setup.

How people lose this round

• It's 36.A bare number, right or wrong, shows nothing. They're grading the route, so say it as you go.
• Sorry, I mean a half?A new number with no reason looks like guessing. Check first, then change the answer.
• Four rolls at a sixth each, so two thirds.Adding chances counts the overlaps twice. For "at least one", take one minus the chance of none.
• Give me a few minutes to think.Long silence gives them nothing to judge and no way to help. Think out loud, even while you're stuck.
• No, I think my way works.Brushing off a hint costs more than needing one. Taking it well is part of what they're watching.

Stage by stage

Close each stage out loud, so they always know where you are.

1. 1The situation in one line15 secThat's the situation. Here's what I did.
2. 2What you did60 secThat was my part. Here's how it turned out.
3. 3The result, with a number30 secThat's the result. And here's what I'd change.
4. 4What you'd change15 secThat's what I'd do differently. Happy to go into any part.

How people lose this round

• We decided, we built it, and we fixed it.They're hiring you. With "we" all the way through, they can't tell what you did.
• I can't think of a real failure.Everyone has one. It sounds like you don't notice your mistakes, or won't own them.
• In the end, they saw I was right.A story you win outright sounds one-sided. Show what you took from the other side.
• So to give you some background, the company had three teams...Long setup uses up your time and their attention. One line of situation, then what you did.
• It worked out, so it was the right call.A good result can come from luck. Say why the decision made sense with what you knew then.

Stage by stage

Close each stage out loud, so they always know where you are.

1. 1The problem2 minThat's the problem and why it mattered. Here's the main choice I made.
2. 2Your choice and its alternative3 minThat's what I chose and what I turned down. Now what broke.
3. 3What broke2 minThat's what went wrong and how I found it. Here's where it ended up.
4. 4The result2 minThat's the result, against where we started. Last, what I'd do next.
5. 5What's next1 minThat's what I'd do next. Happy to go further on any part.

How people lose this round

• The stack was Python, Postgres, Airflow and Docker.A list of tools isn't a decision. Say what you chose, what you turned down, and why.
• Nothing really went wrong.Every real project breaks somewhere. It sounds like you didn't build much of it, or didn't watch it run.
• We got 95 percent accuracy.Without the baseline or how rare the positives are, the number means nothing. Say what the simple approach scored.
• I tried logistic regression, then trees, then a neural net.A list of tries isn't a story. Say what each one taught you and why you moved on.
• The team built it, and I helped out.Too modest to score. Name the parts that were yours.