Bloomberg New Grad Software Engineer Interview (London, 2019): Phone Screen to a Single Onsite Round
Updated July 17, 2026
A candidate applied for a new-grad software engineer role at Bloomberg's London office and went through a phone interview followed by a single onsite round. The phone screen opened with a walkthrough of the candidate's CV and one academic project, then moved into a technical question built around matching parenthesis-like characters in a string, with several follow-ups that pushed the solution toward a more general form.
About 24 hours after the phone interview, the candidate was invited onsite, which took place roughly a month later in Bloomberg's London office and was expected to run across two days. Only one onsite interview happened, a panel of three engineers focused on a custom string-sorting problem. No second interview followed. After a wait of about a day, a rejection email arrived a few days after the onsite, with no detailed feedback.
How the process went
CV Screen / Phone Interview Intro
The phone interview opened with a review of the candidate's CV and a discussion of one project from their studies before technical questions began.
Phone Interview - Technical
A roughly one-hour technical discussion centered on a generalized parenthesis-matching problem, with several follow-up extensions; the full call ran about 1 hour 2 minutes.
Phone Interview Result
The candidate was invited to an onsite interview within about 24 hours of the phone call.
Onsite Scheduling
The onsite took place roughly one month later at Bloomberg's London office and was expected to span two days.
Onsite - Round 1
A single technical interview with a panel of three engineers, centered on a custom string-sorting problem based on the Welsh alphabet.
Post-Onsite Wait & Outcome
No further onsite interviews followed. After about a day of waiting, a rejection email arrived roughly three days after the onsite interview, with no detailed feedback given.
Phone Interview
Data structures and generalizing a stack-based matching problem · ~1h 2m
- Walk through your CV and describe one project from your studies.
- Modified version of Valid Parentheses: determine whether a general string, not limited to bracket characters, has valid matching pairs, using a stack.
- Follow-up: generalize the solution to handle many different open/close character pairs (e.g., "<>", "aA") without writing a separate check for each pair type.
- Follow-up: reduce the two lookup arrays used in the generalized solution down to a single array.
- Follow-up: state the time and space complexity, and describe the input that produces the worst-case O(N) space.
- Follow-up: how would you adapt the solution if the stack had a fixed capacity smaller than the input string length?
The interviewer indicated he was less concerned with a fully bug-free implementation than with whether the candidate understood the underlying concept.
The candidate was invited to ask one question about the role at the end of the call.
Onsite Interview (Panel of 3)
Custom sorting logic built on a non-standard alphabet
- Given the letters of the Welsh alphabet, including two-character letters such as "ch", "dd", "ng", "ll", "ph", "rh", and "th", and a list of strings, sort the strings according to that alphabet's order.
- Follow-up: how would you test the code?
- Follow-up: how would the approach change if the strings also contained capital letters?
One of the three engineers led the conversation while the other two mostly listened and asked occasional questions.
The candidate solved the problem with some guidance from the interviewer.
This was the only onsite interview the candidate completed; a second day of onsite interviews that had been expected did not take place.
Key takeaways
- Practice adapting canonical problems, like bracket matching, to generalized versions; interviewers may extend a familiar prompt to test whether you can abstract the underlying pattern rather than recall one memorized solution.
- When a follow-up asks you to generalize a solution, look for approaches (such as array or hashmap lookups keyed by character) that scale without adding a new branch or helper function per case.
- Be ready to reason about complexity trade-offs on the spot, including identifying which input produces the worst-case space usage.
- Prepare a specific, researched question to ask your interviewer at the end of a call or round; it can leave a stronger final impression than a generic question.
- A multi-day onsite loop is not a guarantee that every scheduled interview will happen; some processes end early without much explanation, so treat the schedule itself as weak evidence about the eventual outcome.
Practice a Bloomberg interview
Rehearse out loud against the kinds of questions in this story — with an AI interviewer that asks follow-ups.
Practice this interviewSource
The questions and process facts come from the candidate's public write-up, linked below. The retelling above is our own summary.
Candidate's public write-up on LeetCode Discuss