I finished a full loop at Two Sigma a few weeks ago. The behavioral portion was more substantive than I expected from a quant firm, so I wanted to share what actually came up.
Quick context: I was interviewing for a senior engineering role after about 18 months away from full-time work. I was a little anxious about the behavioral round assuming they'd be dismissive of a gap. They weren't, but they did ask about it directly and clearly.
Questions I was asked, paraphrasing: Walk me through a time you had to make a decision under significant time pressure with incomplete data. What did you choose, and what happened? Tell me about a situation where you disagreed with the technical direction your team was taking. How did you handle it? Describe a project where the scope changed significantly mid-way. What did you do? How do you handle situations where you need to say no to a stakeholder request?
The through-line I noticed: they want to see intellectual confidence without arrogance. Especially how you handle uncertainty and disagreement. The "incomplete data" question kept coming up in different forms across two rounds.
For the "disagreed with direction" question, I told a story about pushing back on a caching approach that turned out to be right. The interviewer asked me three follow-up questions. They weren't trying to trip me up. It felt like they were curious whether I could sustain a reasoned position under mild challenge.
They did NOT ask typical STAR-format prompts that bottomed out in easy wins. They probed the messy middle. What did you consider, what did you get wrong, what did the situation teach you.
One thing that stood out: the interviewer explicitly said they value people who "disagree and commit" but also people who don't commit until they've actually worked through the disagreement. Make of that what you will.
For prep, I'd treat it like any serious behavioral round but lean into the nuance of your stories. Easy wins with a happy ending read as surface-level there.