[New Grad] Hyundai Motor — CarCloud Role — Coding Test + 1st Interview (Rejected) Recap (2023 H1)

I never posted about my Hyundai Motor CarCloud-role coding test and 1st interview before — sharing the recap now.

I applied for Hyundai Motor's 2023 H1 new-grad recruitment. I was rejected at the 1st interview, but the content was rich enough to be worth a writeup.

The 2023 H1 Hyundai Motor CarCloud role timeline (through the 1st interview) — for your reference.

This post focuses on the AI interview / aptitude, coding test, and 1st interview after the document screening passed.

Figure 1. Document screening — passed

Aptitude + AI Interview

Aptitude + AI interview window: 2023-04-04 (Tue) 07:00 – 22:00

You took the aptitude test and AI interview anywhere between 07:00 and 22:00 that day. I was working at the time, so I took a half-day off in the afternoon to do mine.

A. Aptitude test (~1 hour)

Standard personality assessment. About 1 hour. Just answer consistently with yourself. Nothing too tricky — don't stress.

B. AI interview — ViewInter platform (12 questions total)

Reconstructed from memory and what I jotted down right after.

Each item gives 1 to 1.5 minutes to answer, with 30 seconds to 1 minute of prep time after seeing the prompt. (Total ~30–40 minutes)

1. Self-introduction and motivation
2. Work-related experience relevant to the job posting
3. Question about Hyundai Motor's products

Combo questions — 3 questions each across 3 sets. I didn't record all questions, so here are the themes:

(Set 1) Setting and pursuing a shared goal with teammates - Questions about team collaboration experience, tackling something hard together, and how I overcame the painful parts. (Split into 3.)

(Set 2) Ideas-related theme - Coming up with a creative idea, struggles in the process, opinion clashes with people around new ideas. (Split into 3.)

(Set 3) Resolving opinion clashes with teammates - Experiences where opinions differed during collaboration, how I overcame difficulties, what I felt about it. (Split into 3.)

Figure 2. AI interview + aptitude — passed

AI interview + aptitude result came in 3 days. Announced on 2023-04-07.

Coding Test

Coding test: 2023-04-18 (10:00 – 12:00, 2 hours)

3 problems total in 2 hours. I solved problems 1 and 2; problem 3 I never got to read, so I don't remember it.

Coding test platform — Softeer

Softeer — Hyundai Motor Group SW Talent PlatformSofteer is for software Engineer.softeer.ai

📝 I remember the problems enough to reconstruct, but I never reuploaded my code to GitHub. So I'll describe them roughly.

⚠️ I'm writing from memory — these are approximations, not the exact problems.

Problem 1. Implementation (Simulation)

Hand-drawn reconstruction from memory

Numbers laid out on a grid as in the figure, with rules where the top-left is (1,1) and the bottom-right is (n,n) — given (r, c), return the number written at that coordinate. e.g., given (3, 4), return 12.

Problem 1 thoughts: not hard, but under the pressure of 3 problems in 2 hours, the pattern didn't jump out at me 😢 Took me about 50 minutes.

Problem 2. Permutation / Combination (Math)

Hand-drawn reconstruction from memory

Given N people and M events (e.g., N = 6, M = 3, with N ≤ 20), and a skill matrix showing each person's skill level for each event.

To make the sports day fun, you want to form two teams such that the per-event skill gap is minimized. Teams must be split exactly in half (people count is even, fortunately).

My approach:

→ Choose half via combinations, compute per-event skill-gap (absolute value) across the two teams, and return the minimum total gap.

💡 Tip: Permutations would also give the right answer, but with N = 20, permutations means O(20!) — TLE. You have to use combinations to fit in the time limit.

Problem 2 thoughts: I solved in Java, so no Python itertools — I had to implement Combination / Permutation from scratch, which ate time 😢 I also took 5–10 minutes to recognize it as a combination problem, so the total ran about 1 hour.

Problem 3 — don't remember

Didn't even get to read problem 3 — only solved through problem 2.

CodingTestReproduction/HyundaiMotor-codingTest · taehyuklee/AlgorithmRepo tracking problems solved on online judges (Baekjoon · Programmers · Softeer · CodeTree). Notes in README.md.github.com

1st Interview

The 1st interview = pre-assignment + technical interview.

🔒 I can't reveal the specifics of the pre-assignment topic. Hyundai explicitly asked us to keep it confidential, so I'll only cover the format, how the technical interview handles the pre-assignment, and the overall procedure.

1. Pre-assignment

The pre-assignment is what you solve and then present (5–10 min) to the interviewers when you enter the technical interview. You log in to the pre-assignment platform 40 minutes before the technical interview and wait for 10 minutes in an online waiting room. Then a proctor (an actual person joining the room online) guides you, and you spend 30 minutes reading the problem and writing your solution in an online editor.

📝 The online editor supports rich formatting — tables, headings, font weight, color, etc.

I can't go into specifics, but at a high level: the problem is relevant to the role you applied for. It's the kind of problem you'd actually face on the job. I can only speak from my role's experience, so other roles may differ.

How I prepared (my approach)

This is probably the part you're most curious about. The pre-assignment format was first introduced in 2023 H1 (apparently it existed long ago, but had been absent recently). So I prepared based on rough predictions about what would come — and my predictions turned out roughly right.

"The CarCloud role probably runs a large-scale service with massive data, so server development and infra-side problems would come up."

Working under that assumption, I thought through overall architecture, server development methodology, etc. At the time I was actually working on an API Gateway solution project with real high-traffic exposure — lucky for me.

2. Technical interview (3 interviewers vs. 1 candidate)

After the pre-assignment, the proctor guided me to the online (video) technical interview room. I gave a short self-introduction and then briefed the pre-assignment. The interviewers mostly checked whether I grasped the problem's intent and whether my solution was logically sound.

You do need some technical depth to even solve it, but they seemed less focused on the "right answer" and more on logical reasoning.

[Interviewer feedback on the pre-assignment]

I defended the pre-assignment well, and the interviewers told me I'd understood the problem well and gotten close to the right answer — which felt good.

After the pre-assignment briefing and defense, they immediately moved to verifying my prior projects. They asked a lot of technical questions; here are some I remember. They were all based on my API Gateway solution project.

1. What authentication/authorization did you use in the API Gateway?

2. You must have used an encryption algorithm — which one?

3. How did you build the API Gateway to handle large amounts of data? - 3.1) What async processing approach did you use? (follow-up to 3) - 3.2) You mentioned Non-Blocking — how did you implement it, and what is Non-Blocking? (follow-up to 3.1)

4. You mentioned using Swagger on the Java side for Open API — which version? How did you use it?

5. To handle high traffic, what does the infra need to look like, and which cross-functional teams need to collaborate?

6. You seem solid on Java — have you used other languages? - 6.1) Do you know the difference between Python's Thread and Java's Thread?

That covered the major questions I remember. Out of 6 big questions, I answered 4 well, 1 wrong, 1 "don't know" — when I self-graded, I thought "the pre-assignment was solid and the resume verification batting average is decent, so the odds are good."

7. Anything to say at the end?

But... I got rejected 😢

1st-round result on 2023-05-12, announced around 11:25 (oddly enough, the date I'm posting this is exactly one year later).

I checked the result during lunch at work — too much of a shock that I didn't save the screen (I usually save rejection screens too). (That evening I went out with my coworkers for KyoChon chicken and had a highball.)

Wrap-up

That ends my Hyundai Motor journey. Preparing was tough, but it gave me a chance to reflect on myself, and there were plenty of fun moments too.

Hope this post helps someone out there :)