Summary
- The Transition: Balancing a Mechanical Engineering major with Software Engineering, managing the massive workload of two demanding disciplines.
- Rapid Adaptation: Restarted algorithm practice in February; passed the coding test for Hana TI (Hana Bank’s IT subsidiary) in March and secured my current role at KT DS(Korea Telecom’s IT subsidiary) by April.
- The Learning Curve: From receiving blunt peer feedback after crying in my first interview to navigating GM, AWS, and high-growth startups while starting a new job.
- Strategic Choice: Staying at my current KT subsidiary role to leverage enterprise-scale AI development and a substantial compensation gap for technical compounding.
A Year of Expanding Horizons
2024 was the year I took the first real step into my career. For the first time in my life, I committed to the job-seeking process. After graduation, I had spent time avoiding the pressure of the market, but in February, I decided to face it head-on.
The momentum was unexpected. I picked up algorithm practice again in February. By March, I passed the coding test for Hana TI(Hana Bank’s IT subsidiary), and by April, I passed the test for my current company, KT DS(Korea Telecom’s IT subsidiary). This period was a crash course in how different organizations evaluate engineers.
The First Interview: Tears and Lessons at Hana TI
My first-ever interview was at Hana TI (Hana Bank’s IT subsidiary). I arrived at their training center in Cheongra, feeling incredibly nervous. Everyone was in full suits, and I could sense a high level of competition.
When it was my turn, my voice shook during the formal greetings. My mind went blank, and I couldn't even introduce my project properly. But what really broke me was the gap between me and the other candidates. They were all talking about their prestigious external experiences—Naver Boostcamp, Kakao Bootcamp, Samsung Bootcamp(SSAFY).
Compared to them, I felt like a complete amateur. I had nothing but my college graduation on my resume; no fancy boot camps, no high-profile external activities. I was so intimidated and felt so small that during the final remark, I actually started crying.
The interviewers were so kind that they even apologized, asking me, "Was this too much of a stress interview for you?" I felt so embarrassed because the atmosphere wasn't stressful at all; it was my own lack of confidence that overwhelmed me. After the interview, I exchanged numbers with another candidate for the feedback. Their feedback was blunt: "Your technical answers were logical, but you didn't need to apologize for your lack of experience." That day taught me to stop apologizing for my path and start owning my logic.
The Small SI: "We Need an Interpreter"
Shortly after, a small SI firm contacted me for my English scores. They proposed a hybrid role: backend development combined with interpretation for their Indian developers. I told them I didn't want to interpret, so they set up a developer interview instead. They offered me 28 million KRW ($21k). I declined, but getting a final offer helped rebuild the self-esteem I had lost in Cheongra.
Joining the Platform Division at KT DS
the process was a grueling marathon: Document Screening → Personality Test + Aptitude Test + Coding Test → PT Task Preparation → PT & Technical Interview → Health Check → Executive Interview.
The coding test and aptitude test were the first hurdles. The aptitude test was particularly demanding, featuring complex problem-solving taskst. For the PT interview, I had to create a presentation on a given topic in just one hour without any outside resources, all while being monitored on camera. I studied twelve potential topics in advance, and after the task, I practiced my five-minute speech until I had the script completely memorized.
In the first interview, I was so determined not to cry again. When the interviewers asked for a final remark, I blurted out, "I won't be giving a final remark because I cried while giving one in my last interview." The interviewers actually burst into laughter at my blunt honesty. Despite my voice shaking like a goat throughout the technical questions, I passed. For the executive round, I even stayed at a hotel near the site to ensure I was in peak condition. Despite my voice shaking like a goat, I passed. For the executive round, I even stayed at a hotel near the site to ensure I was in peak condition.
Then came the most anxiety-inducing period: The Placement. During the entire onboarding process, I was terrified. I constantly prayed that I wouldn't be sent to a team doing only legacy maintenance or, worse, be dispatched to a remote third-party client site as an SI developer. I wanted to build something modern; I wanted to grow.
My prayers were answered. I was assigned to the AI Team under the AI Department of the Platform Division. I am now part of the team developing AI Contact Centers (AICC)—including chatbots and voicebots—and backend systems for AI Speakers. Because we are a KT subsidiary, this is effectively in-house development for massive infrastructure.
General Motors (GM) - Infotainment Backend
While I was going through the final stages for KT DS, I was also in the middle of the General Motors (GM) recruitment process. I passed the phone screening, the coding test, and the AI English interview for their infotainment backend role. However, the final executive round in July collided with my onboarding orientation at KT DS. I was too worried about missing orientation and didn't think to ask for flexibility. In hindsight, I was following the rules too strictly. Today, I would use my leave to protect that high-leverage moment.
General Motors (GM) - Infotainment Backend
While I was going through the final stages for KT DS, I was also in the middle of the General Motors (GM) recruitment process. I passed the phone screening, the coding test, and the AI English interview for their infotainment backend role. However, the final executive round in July collided with my onboarding orientation at KT DS. I was too worried about missing orientation and didn't think to ask for flexibility. In hindsight, I was following the rules too strictly. Today, I would use my leave to protect that high-leverage moment.
AWS Assessment Experience: The Role-Playing Simulation
AWS used a different filter. I received a coding assessment followed by behavioral simulations modeled after real production incidents.
The algorithm portion had three problems. I solved two but ran out of time on the last one, which required Binary Search. I took the test on a weekday after a 90-minute commute, feeling absolutely exhausted. The behavioral section—making decisions about debugging outages under pressure—was the most interesting part. It changed how I prepare for high-stakes interviews. I learned to treat execution conditions—like energy and environment—as part of the preparation itself. Failure, when read carefully, is just data.
What stood out was the behavioral simulation. It felt like a role-playing game where I had to choose options to debug errors encountered during work. It mirrored the reality of production incidents far more than standard coding tests.
Startup Interview #1: Red Flags and the Practical Gap
I had applied to this startup before joining my current company. I passed their coding test first, and the interview was scheduled right during my onboarding period at KT DS.
This company raised several red flags: the pay was extremely low, and after researching their financial status, I found they were in a very unstable position with high job insecurity. During the technical round, they asked why I solved a specific coding problem the way I did. I answered, "In my experience, PriorityQueue was always faster, so I used it." They immediately followed up: "Why is it fast? How is it implemented internally?" I couldn't answer. I knew it used a Heap, but I couldn't explain the mechanics under pressure. Even though the company itself wasn't the right fit, that moment was a wake-up call. I realized I was just "writing code" without knowing the "why."
Startup Interview #2: Reasoning Over Rote Memory
The startup pipelines became the intellectual center of my year. The most notable was an interview with an industry-leading startup (#1 in its sector) with over 1 million MAU. While the coding tests were manageable, the interview itself shifted from practical application to fundamental principles.
When the topic of Garbage Collection (GC) came up, the CTO asked a striking question: "How do you think the system actually knows that an object is alive?"
Since I didn’t have the textbook answer, she encouraged me to deduce it. I reasoned that objects might maintain a list of what is currently using them—if that list isn't empty, the object is alive. She challenged this, pointing out the massive inefficiency. I pivoted my logic and proposed Reference Counting: incrementing a counter when a reference is created and decrementing it when one is destroyed. If the count hits zero, the object is dead.
She nodded and said, "There are indeed systems that work that way." Throughout the discussion, she observed my thinking process and led me toward conclusions through subtle hints. It was exhilarating to be pushed to reason from first principles rather than just reciting from memory. I was deeply impressed by a culture that valued raw deduction over rote recall.
The Offer and the Hardest Calculation
I later heard that both the CTO and the Backend Lead were very impressed with my performance. However, when the offer arrived, it sparked a week of intense agony. Because of my financial situation, I had a firm baseline, and the offer was several million KRW below that.
The financial gap was substantial—nearly 18 million KRW (approx. $13,500) per year from current salary. But beyond the money, I was paralyzed by a much darker fear. I was battling severe imposter syndrome, fueled by the fact that I had to split my focus between Mechanical Engineering and Software Engineering for years.
I was terrified. I kept thinking: "If I leave this 'greenhouse' now, and the startup fails or I get fired because I can't keep up, will I even be able to get another job?" At that moment, I genuinely believed that my skills weren't good enough to pass another interview at a different company. I felt like my current employment was a stroke of luck that might not happen twice. The thought of being cast out into a freezing job market without a safety net was bone-chilling.
My family, seeing me struggle from the bottom up—from Liberal Arts to Mechanical Engineering, and then to Software Engineering—shared this fear. They were finally relieved to see me in a stable "open recruitment" role and were terrified I would "throw it all away" only to be crushed by the reality of a global recession. Every morning, the links to recession news in our family group chat felt like a warning of the doom awaiting me if I made the wrong choice.
Ultimately, my anxiety won over my excitement. I made the strategic decision to stay. Not to hide, but to build the inner strength and technical depth—mastering Effective Java, SQL Tuning, and System Architecture—until I no longer fear the wild. I chose to wait until my confidence is rooted in real mastery, not just luck.
2025: Forward, Deliberately
2024 was about friction and diagnosis.
- Hana TI taught me to stop apologizing and show my logic.
- GM & AWS taught me to protect my energy and timing.
- The Startups taught me how to think from first principles.
- KT DS taught me the value of in-house enterprise AI development.
I stayed in my greenhouse to grow roots strong enough for the next leap. 2025 will be about building the depth to match my ambitions.
Top comments (0)