JP Morgan Code for Good interview: process, rounds and questions

Written by Mentari as typical of this process — not a record of questions JP Morgan has asked. Each comes with the follow-up a good interviewer asks next and what they listen for in the answer. Scenario details are our framing, chosen to fit JP Morgan’s sector.

1. 01 Mirror a binary tree, then tell me how you would check two trees are mirrors of each other.

   Follow-up · The trees have a million nodes and you are worried about recursion depth. What changes?

   Listens for · Correct recursion, then an iterative or explicit-stack alternative with a reason.

2. 02 Model a lift system as classes — no code, just the design.

   Follow-up · Now there are three lifts and one call button per floor. What changes, and what stays?

   Listens for · Abstraction chosen before implementation — the OOP-design probe candidates report.

3. 03 Partition a string so every piece is a palindrome, minimising the number of cuts.

   Follow-up · Which subproblem did you memoise, and what would it cost not to?

   Listens for · DP state identified clearly; complexity stated.

Compiled from 10 candidate reports (2024–2026) and the official hiring page. Where reports differ, the range is shown rather than one number.

HackerRank — reported 2025 pattern — Two coding problems, 60 minutes in one 2025 Code for Good report and 90 minutes in a Software Engineer FTE report; no aptitude, verbal or numerical sections. Difficulty is reported as easy to medium with occasional DP or graph problems.^5,6,14

1. Work through a pair of array and string problems, easy to medium, inside an hour, then do the same with a DP and a graph problem. The HackerRank round is two problems in 60–90 minutes; some candidates get DP or graphs.
2. Record yourself answering "tell me about a difficulty you overcame" and "a time your team disagreed" in two minutes each, with 30 seconds to think. HireVue uses exactly that format and is reported as the round that eliminates most.
3. Build something small in a team over a weekend and present it in five minutes. Code for Good is a team build-and-present event; practising the presentation matters as much as the code.
4. For a direct drive: prepare trees, DP and implementation on paper, and your CV to two layers of depth. The technical panels in direct drives examine the CV closely and set coding on a call.
5. Read JPMC’s description of the Software Engineer Program and be able to say what "business analysis" means in a bank. The official profile names it; a candidate who treats the role as pure coding misreads the programme.
6. Keep a few curiosity answers ready — the last book you read, a probability puzzle. The senior conversation in direct drives is reported to go there.

What JP Morgan says it assesses

• A four-stage journey — explore, apply, interview, decision — in which "the interview process will differ for each position" and candidates "can expect to meet with several people in multiple rounds"¹
• Code for Good: "showcase your tech skills and work alongside a team, guided by our technologists"²
• Software Engineer Program: foundational programming skills plus an understanding of business analysis, development, maintenance and software improvement⁴

What candidates report being probed on

• Recorded behavioural answers — a difficulty overcome, a team conflict, something learned — under a two-minute clock^5,6
• Easy-to-medium coding under time: arrays, strings, next permutation, reachability^5,7
• Direct drives: trees, DP, implementation on paper, the CV line by line^10,8
• Java and OOP design — a lift modelled as classes, no code — and exception handling across an inheritance chain¹¹
• A senior interviewer’s curiosity questions — maths, probability, what you read¹⁰

JPMorganChase is the only major recruiter in these guides whose main India campus route is a hackathon rather than an interview. Code for Good puts shortlisted students into teams building for a nonprofit over a weekend, with JPMC engineers mentoring, and offers follow from what the teams build and how they present it. The stages before it are filters — a two-problem coding test and a ten-minute recorded behavioural interview — and candidates single out the recorded round as the one that eliminates most, because two minutes on camera about a failure is a skill most engineering students have never practised.

Where the firm visits a campus directly for software engineers, the process looks like a conventional bank-engineering loop: two or three technical rounds, coding on a call, Java and OOP design, the CV examined down to GitHub, and an HR or senior conversation that can wander into probability and books. The two routes are selected for by different things — teamwork and shipping under a deadline in one, individual problem-solving depth in the other — and the official pages do not say which a given campus will see.

Where the sources disagree

The official pages supply the programme definitions and a deliberately generic journey; the campus accounts supply the mechanics, and they diverge by route rather than contradict each other. The Code for Good accounts from 2025 agree on shortlist, HackerRank, HireVue, hackathon, and differ only on test length (60 versus 90 minutes). The direct-drive accounts describe technical interviews: three rounds with five coding problems (July 2024), separate technical and HR rounds (August 2025), a software-profile drive that tested 400, interviewed 88 and selected 32 through three technical rounds and HR, and an IIT Roorkee internship drive with a 65-minute two-problem test, one technical interview, a business problem-solving round and HR under a 7.0 CGPA cut-off; an older account adds two-interviewer panels and a VP conversation. On eligibility there is nothing official beyond "enrolled in a bachelor’s degree program"; the 8.5 CGPA figure is an observation about who was shortlisted at one campus and should be read that way.

4. 04 Walk me through the most important line on your CV — the one you would defend hardest.

   Follow-up · Now the one you would delete if you had to. Why that one?

   Listens for · Ownership and honesty about the thin spot.

5. 05 Your project used a machine-learning model. What is its false-positive rate, and what did a false positive cost?

   Follow-up · In a bank, which error is more expensive — a false positive or a false negative — and does your model know that?

   Listens for · Metrics connected to cost; awareness that the domain sets the trade-off.

6. 06 A charity you have never heard of needs a volunteer-scheduling tool by Sunday night. What do you build first?

   Follow-up · Your teammate wants to add logins and roles. Do you let them?

   Listens for · Scope discipline — the Code for Good instinct — and a reason for the call.

7. 07 Tell me about a time your team disagreed about what to build.

   Follow-up · What did the person who lost the argument do next, and what did you do for them?

   Listens for · A real disagreement and care for the person on the other side.

8. 08 Two dice: what is the probability the sum is seven? Now, given it is odd?

   Follow-up · Explain the second answer to someone who has not studied probability.

   Listens for · Correct conditioning, then plain-language explanation — the senior-conversation style candidates report.

9. 09 What did you last read that was not for a course, and what did it change?

   Follow-up · Would you recommend it to me? Why or why not?

   Listens for · Curiosity and a view; not a prepared title.

10. 10 Why a bank rather than a product company for your first engineering job?

    Follow-up · What do you expect to be slower here, and will you tolerate it?

    Listens for · A realistic reason and an honest answer about constraints.

In two minutes: tell me about a time you failed at something and what you learned.

Weak · "In my second year I failed to complete a project on time because of poor planning. I learned the importance of time management and now I always plan ahead and meet my deadlines."

Stronger · "Our team’s hackathon demo crashed on stage because I had merged my part twenty minutes before without testing against the others’ changes. We placed last. What I changed is specific: I now keep a working build at every merge and demo from that, not from my branch. In the next event we won second — the build was uglier and it ran."

Why the follow-up lands · A live interviewer later asks "what did your teammates say to you after the crash?" The strong story has that moment; the weak one has "time management".

Illustrative — yours must be true.

JPMC’s Code for Good route is a two-problem coding test, a ten-minute recorded behavioural interview and a weekend hackathon; direct campus drives use technical panels and a senior conversation. What a cell can train is coding under the clock and answering on camera — the HireVue round is where most eliminations are reported.

Mentari’s software-engineer mock rehearses the live technical conversation — project depth, a problem reasoned aloud, a behavioural story with follow-ups — and its text mode doubles as timed-answer practice for the recorded round. Run JP Morgan-pattern mocks for a whole batch.

• Software engineer mock interview
• HR round mock interview for campus placements
• Goldman Sachs interview
• Morgan Stanley interview
• STAR Method Answers That Survive Follow-ups
• Tell Me About Yourself (Freshers) + 3 Follow-ups
• All company guides →

1. How we hire — JPMorganChase (official) · official · accessed 2026-10-23
2. Tech for Social Good hackathons — Code for Good — JPMorganChase (official) · official · accessed 2026-10-23
3. Code for Good 2025 — candidate agreement and rules — JPMorganChase (official) · official · accessed 2026-10-23
4. Software Engineer Program — full time — JPMorganChase (official) · official · accessed 2026-10-23
5. JPMorgan Chase internship interview experience — Code for Good 2025 (pool campus) — GeeksforGeeks · candidate report, 2025-03 · accessed 2026-10-23 · shortlist → HackerRank (60 min) → HireVue → hackathon
6. JPMorgan Chase Code for Good interview experience — Software Engineer FTE — GeeksforGeeks · candidate report, 2025-06 · accessed 2026-10-23 · HackerRank (90 min) → HireVue → hackathon from 17 June 2025
7. JPMorgan Chase interview experience — on-campus (Code for Good 2024) — GeeksforGeeks · candidate report, 2024-08 · accessed 2026-10-23 · coding test questions; partial account
8. JP Morgan interview experience — on-campus, July 2024 (Software Engineer) — Naukri Code360 · candidate report, 2024-07 · accessed 2026-10-23 · three rounds; five coding problems in a video-call technical round
9. JPMorgan Chase software engineer internship — India, August 2025 — Taro · candidate report, 2025-08 · accessed 2026-10-23
10. JPMorgan interview experience — SDE I — GeeksforGeeks · candidate report, 2024-04 · accessed 2026-10-23 · older account: two technical panels and a VP conversation; cited for the senior-conversation probe only
11. JP Morgan (software profile) on-campus interview experience — GeeksforGeeks · candidate report, 2025-07 · accessed 2026-10-23 · GfG page last updated July 2025, drive undated; 400 tested, 88 interviewed, 32 selected; three technical rounds and HR
12. JP Morgan Chase summer internship interview experience — IIT Roorkee — GeeksforGeeks · candidate report, 2025-07 · accessed 2026-10-23 · GfG page last updated July 2025, drive undated; 65-minute two-problem test; 7.0 CGPA cut-off
13. JP Morgan interview experience — Software Engineer Program (HireVue) — GeeksforGeeks · candidate report, 2025-07 · accessed 2026-10-23 · HireVue technical-plus-video screen
14. JP Morgan coding questions — PrepInsta · candidate report · accessed 2026-10-23 · preparation-site compilation

Verified October 2026 by Sumit Mishra, Mentari founder. Spotted an error, or represent JP Morgan? Write to [email protected] — we acknowledge within two working days and correct or remove a disputed statement within 72 hours.

Mentari is not affiliated with JP Morgan. This page is pattern-based, compiled from the public sources listed, and is not insider information. Candidate-reported details are reproduced as patterns only, in our own words.