Bloomberg interview notes, page 4 of 14

Large Enterprises

Most of these Bloomberg write-ups come from Software Engineers (130), then Product Managers (2) and Data Analysts (1). The most reported rounds are the technical phone screen (100), the behavioral / leadership round (8), the onsite coding round (8) and the system design round (8).

133 interview notes · updated Oct 2026

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with an interactive coding problem that punishes brute force pretty hard. The API call limit forces you toward a smarter approach, which was a nice change from the usual grind-it-out problems.

Given a rectangle defined by its top-right and bottom-left corners, count the number of ships inside it. You can only query a helper function that returns true if at least one ship exists in a given sub-rectangle. There are at most 10 ships total, and you're limited to 400 calls.

“The call limit is what makes this interesting.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a classic streaming data problem. Not much to say about the setup but the question itself kept me busy.

Given a continuous stream of numbers, design a data structure that can return the median at any point in time.

“Two heaps.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a linked list problem that looks manageable until you actually try to get the pointer reconnection right. The recursive child-flattening part is where things get fiddly.

Given a doubly linked list where nodes can have a child pointer referencing another doubly linked list (nested arbitrarily deep), flatten it into a single-level doubly linked list. Child lists should be inserted immediately after the node that holds them, and all child pointers must be nulled out with prev/next pointers correctly updated throughout.

“I got the high-level idea pretty fast: depth-first, splice the child in, move on.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a classic greedy algorithm problem. Nothing too wild, but you do need to actually justify the approach and not just code it up.

Given 2N candidates each with a cost to fly to either New York or San Francisco, how do you assign exactly N to each city to minimize total travel cost?

“The solution itself isn't hard to land on.”

View Post

Bloomberg·Product Manager·Onsite - Behavioral / Leadership

May 2026

Bloomberg PM interview, behavioral round focused on product leadership background. One big open-ended question about your full product history, which sounds easy until you're actually in it trying to hit roadmap ownership, global launches, cross-functional work, and a career pivot all in one answer.

Walk me through your product management background: a product you owned end to end, how you handled the roadmap, any global launches, cross-functional work, and how your financial services background set you up for this role.

“This is one of those questions that feels like a softball until you realize you're being asked five things at once.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a classic but tricky parentheses problem. The question had some depth to it once you got past the surface reading.

Given a string with parentheses and other characters, remove the fewest parentheses possible to make it valid. Return all unique valid results.

“I went with BFS first because it felt safer to explain level by level, but the interviewer pushed me on efficiency pretty quickly.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE coding round, one main problem the whole time with a bunch of follow-ups layered on. The core question wasn't too bad but the follow-ups kept coming and I definitely stumbled on a few of them.

Design a class that tracks user activity logs. It should support an add(timestamp, userId) method and a getActiveUsers(currentTimestamp) method, where an active user is one who has appeared in the last 5 minutes.
How would you modify the solution if timestamps are not guaranteed to be monotonically increasing?
How would you scale this system to handle a very large volume of log data?
If getActiveUsers is called very frequently, how would you optimize for that access pattern?

“My first instinct was queue plus hashmap with reference counting, which worked fine for the basic case.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a backtracking/DP problem. Nothing too wild but Word Break II is one of those questions that looks manageable until you start thinking about all the edge cases.

Given a string and a dictionary of words, return all possible ways to segment the string into valid dictionary words, where words can be reused.

“Classic backtracking with memoization.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a tree traversal problem that looks straightforward until you work through the column indexing logic. The follow-up was just reversing the output, which felt almost too easy after the main part.

Given a binary tree, perform a traversal and output one character per vertical column. For each column, pick the deepest node; if there's a tie in depth, pick the rightmost one. Concatenate the results left to right by column index.
Follow-up: now output the same result but in reverse column order, right to left.

“The column tracking is the real puzzle here.”

View Post

Bloomberg·Software Engineer·Technical Phone Screen

May 2026

Bloomberg SWE interview with a meaty API design question that was more involved than I expected. The problem looked like a simple buffering exercise but kept expanding the more they probed.

Design a PacketAssembler API that accepts streaming byte fragments and emits a completed payload once the expected number of bytes have been received. The first fragment carries a header byte indicating the total payload size. Walk through class and method signatures, how you manage internal state, behavior across consecutive packets, error handling for bad headers or oversized input, whether state resets after emitting, and thread-safety.

“I started with the happy path and it went fine, but then they asked about consecutive packets and I fumbled a bit.”

View Post