Jane Street Software Engineer interview questions

6 Software Engineers have shared notes from their Jane Street interviews, with the questions they were asked in each round. That is 75% of the 8 Jane Street interview notes on Screna AI.

6 Software Engineer interview notes

Jane Street·Software Engineer·Technical Phone Screen

Jun 2026
Jane Street SWE interview where they asked me to implement a circular buffer from scratch. Pretty focused session, no tricks, just build the thing and make it work.
  • Implement a circular (ring) buffer with full functionality. Performance isn't the focus, just get it working correctly.

“Spent the first few minutes just thinking out loud about the head/tail pointer logic and when to treat the buffer as full vs empty.”

View Post

Jane Street·Software Engineer·Onsite - System Design / Architecture

Jun 2026
System design round at Jane Street for a software engineering role. The whole session was basically one giant question about building a file system from scratch, and they wanted you to go pretty deep on every layer of it.
  • Design a file system service covering the full stack: functional API, data model, storage layout, consistency and concurrency, and scaling from single-node to distributed. What are the key trade-offs in metadata service design?

“This one sprawled in a way I wasn't totally prepared for.”

View Post

Jane Street·Software Engineer·Technical Phone Screen

Jun 2026
Jane Street SWE interview with a stream processing design question. The core problem was interesting but the follow-up about out-of-order records is where things got real. Solid technical depth required throughout.
  • You have a stream transformer that reconstructs a sparse code-time matrix. Input arrives as batches of (timestamp, code, value) tuples, and output should emit one (timestamp, row-of-length-M) pair per completed row, with missing entries filled as -1. Now suppose a small fraction of records arrive slightly out of order. How do you adjust the transformer so the output is still correct?

“This is where I spent most of my mental energy.”

View Post

Jane Street·Software Engineer·Technical Phone Screen

May 2026
Jane Street SWE interview with a pretty gnarly C++ streaming problem. One question, no fluff, just code.
  • Given a stream of batches where each batch is a list of (timestamp, code, value) tuples sorted by time then code, implement a stream transformer that emits pairs of (timestamp, row_vector) where the row vector has one slot per code sorted lexicographically, filled with -1 if that code has no value at that timestamp.

“This one took me a minute to fully parse.”

View Post

Jane Street·Software Engineer·Technical Phone Screen

May 2026
Jane Street SWE interview focused entirely on a two-part system design problem around extending a trade tracking class. The problem kept building on itself and required reasoning through correctness without running any code, which was a different kind of pressure than I expected.
  • You have a Table class that tracks trades and their PnL values over time. It supports adding trades, grouping them via filters, and rendering an aggregated PnL view. How would you redesign view_result() to run in O(1) time by maintaining incremental state, while also correctly handling set_global_filter() which can change the active set of trades at any point?
  • Now assume get_pnl is asynchronous and you want to avoid calling it multiple times for the same trade and time pair. How do you cache these results as class state, and how does that cache interact with set_global_filter() changes, specifically around invalidation and recomputation?

“This took me a while to even frame correctly.”

View Post

Jane Street·Software Engineer·Technical Phone Screen

Apr 2026
Jane Street SWE interview with a pretty involved systems/streaming problem. The core question was about a stream transformer for sparse matrix reconstruction, then they threw in a variant that broke one of the assumptions you'd been relying on the whole time.
  • You have a stream transformer that takes batches of (timestamp, code, value) tuples and reconstructs a sparse code-time matrix, emitting one row per timestamp with missing entries filled as -1. Now suppose the tuples within a batch are no longer sorted by code, only by timestamp. How do you modify the transformer to still emit correct rows, and what are the time and space tradeoffs of different approaches?

“This was the main question and the variant is where it got interesting.”

View Post