Citi Software Engineer interviews often start with a Karat screen on Java fundamentals and debugging, then move to team technical discussions, behavioral questions, and HR. The exact loop varies by team and level.
$118K
Avg. Base Comp
$138K
Avg. Total Comp
2-4 rounds
Typical Rounds
Not reported
Process Length
Many candidates describe a Karat-run first screen as the main early hurdle for Citi software engineers. Several describe Java theory such as equals() and hashCode(), heap and stack memory, garbage collection, or design patterns. They then describe a debugging-style coding task: fixing failing tests, finding a bug in an existing codebase, or implementing two methods after explaining the approach first. One candidate was asked to review a restaurant order-delivery system and make three improvements within an hour. Another spent 10-15 minutes defending an approach, leaving little time to code. Practice reading unfamiliar Java and talking through changes concisely.
Stack-specific questions also appear. Reports mention Spring Boot, Spring MVC, Hibernate, React, Angular lifecycle hooks, SQL, and AWS Availability Zones versus Regions. Senior-title candidates report project architecture deep-dives and system design on microservices, load balancers, and logging. Some reports are Python-leaning, with dictionaries, Git, and API bearer tokens.
Later conversations with team members or managers often cover your projects, teamwork, conflict, and why Citi. Loops and timelines vary by team and level, so confirm the agenda for each round in advance, since one candidate expected a behavioral-only call and found a technical panel.
Synthesized from 63 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Citi process.
I went into Citi's Software Engineer process expecting a technical grind, and the most memorable part turned out to be a pair programming round. I was asked to design an automatic payment system for a parking lot with about 1,000 cars across 10 floors. The system had to support credit cards, debit cards, and every other payment mechanism, so the interviewer wanted to see how I'd structure the gateway layer so new methods could be added without rewriting everything. We worked through it together, and the back-and-forth about classes and responsibilities mattered as much as the final shape of the code.
The rest of the technical side was lighter than I expected. One conversation was a quick check on basics, like whether I knew what a lambda function was, and it moved fast. Another technical session covered Python class questions and microservices, but it felt closer to a behavioral conversation than a deep technical one. The interviewer seemed worn out by then. In one session the next candidate was let in before I had finished, which told me how rushed the day had become.
The last stretch was a behavioral interview with two recruiters. It was straightforward, and the process overall was simple, but it took a while to hear back. I received another offer in the meantime, and I ended up declining Citi's.
Prep tip from this candidate
Practice talking through a design-style coding problem out loud, such as a parking lot payment system that must accept several payment methods, since the pair programming round focused on how you structure the gateway layer. Also review Python class design and microservices basics, because those came up in the technical conversation.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Citi
Write a query to show the number of users, transactions, and total order amount per month in 2020
| Question | |
|---|---|
| Size of Joins | |
| Real-Time Transaction Streaming | |
| Attribution Rules | |
| Dijkstra implementation | |
| Hurdles In Data Projects | |
| Replace Words with Stems | |
| Concurrent LLM Serving | |
| Offer Matching API Design | |
| Loan Model | |
| Seller Type Modeling | |
| String Palindromes | |
| International e-Commerce Warehouse | |
| Text Editor With OOP | |
| Unified Event Pipeline | |
| Client Solution Pushback | |
| Index Fund Return | |
| Above Average Product Prices | |
| Why Do You Want to Work With Us | |
| Your Strengths and Weaknesses | |
| Singly Linked List | |
| Student Tests | |
| Concentric Circles | |
| Measuring Text Difficulty | |
| 2nd Highest Salary | |
| Merge Sorted Lists | |
| Employee Salaries | |
| Subscription Overlap | |
| Closest SAT Scores | |
| Empty Neighborhoods |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report an initial recruiter or HR contact that confirms background, technology comfort, and availability. One candidate says a scheduling mismatch ended their process at this stage, so be ready with clear availability and a short account of your resume projects.
Some candidates report an online assessment before interviews. One describes SQL multiple choice, basic programming, and bash. Another describes two coding problems in 45 minutes plus SQL, React, and Angular question sections. This step does not appear in every report, so it may depend on the team.
Many candidates report a live Karat interview of about an hour. Reported content includes Java fundamentals, design patterns, fixing failing tests, finding bugs in existing code, and implementing a couple of methods. Some report React, Angular, Spring, or SQL questions. Explaining your approach before coding is commonly emphasized.
Candidates report panels or conversations with engineers, tech leads, or managers. Topics include project architecture, Java and Spring Boot, microservices, system design, and cloud basics. Some panels ran longer than scheduled. Expect to connect your own work to the team's needs, and expect variation by team and level.
Candidates report questions on why Citi, conflict, teamwork, leadership, and why they should be hired. Some describe a SuperDay with behavioral and technical interviews, others a department-head conversation and an HR round covering policies. Outcomes in the reports vary, and follow-up speed is inconsistent.
