okay so i applied to jane street as a new grad SWE (not quant research, just SWE) and made it pretty far in the process. didn't get the offer but i learned a ton and wanted to write up what actually helped vs. what i wasted time on.
first: what track are you applying to? there's a difference between the software engineer track and the quantitative researcher track. i was SWE. the quant track is a different beast (probability puzzles, market making, brainteasers). for SWE new grad, it's closer to a rigorous FAANG loop but with some differences.
what they tested: coding: leetcode medium to hard difficulty, mostly graphs, trees, dynamic programming. no "trick" puzzle questions for the SWE track. system design: yes, even for new grad. mine was simplified but not trivial. "design a rate limiter, think about distributed behavior." i was not expecting this as a new grad candidate. some probability/reasoning? i got one mental-math estimation question in a conversation. light, but be able to think out loud about numbers.
what actually helped me prep: grinding the standard stuff (neetcode roadmap, about 120 problems) doing mock system design out loud. not just reading about it reading about what jane street actually does. not to impress them but because some coding questions are loosely finance-flavored being honest when i got stuck. i tried to think out loud and say what i knew and what i didn't. felt like they valued that over silence followed by a perfect answer.
what i wasted time on: answering brainteasers. i spent like two weeks on these because everything i read about jane street mentioned them. for the SWE track in 2025-2026 at least, i didn't get a single classic brainteaser.
timeline: applied october, first contact november, loop finished in january. they move fast once you're in the loop.
if anyone else is prepping for the new grad SWE track, happy to compare notes. this forum thread from last year was the most useful thing i found.