Finding My Way

Today, I’ve decided to start writing blogs again—or perhaps to start properly for the first time. Three years ago, as a freshman at Beijing University of Posts and Telecommunications, I kept a site at momoyeyu.github.io. But those posts were closer to study notes than real writing. So here, on March 31, 2025, I’m starting over.

Why Start Blogging

I’m picking this up because I’m facing a decision I can’t keep deferring. Do I finish my master’s at BUPT and pursue a career as a software engineer? Or do I aim for a PhD at a stronger institution and commit to research? These are fundamentally different paths, and I’ve been oscillating between them for months. Writing helps me think clearly, and I need that clarity now more than ever.

I believe these paths suit different kinds of people, and I’ve spent too long being unsure which kind I am. Recently, though, something closer to an answer has started to take shape. I want to try articulating it here.

Where the Confusion Began

My confusion is rooted in my research experience. Two years ago, I started working as a research assistant under Prof. Jie, who gave me my first real exposure to academic research. Through that work, I met Weifei—one of my closest friends from college. Weifei has genuine passion for research and a relentless drive to excel. Together, we co-authored three papers, one of which was accepted at Usenix Security Symposium 2025, a top-four venue in security. For undergraduates, that’s a meaningful result.

I’m grateful to have contributed to Weifei’s projects, and the publications were valuable for me too. But being a second or third author doesn’t make you a strong researcher on its own. So before the 2025 Spring Festival, I decided to lead my own project, initially targeting InterSpeech. I developed a feasible idea and began working. However, it became clear that a security-oriented paper wouldn’t fit that venue well. We considered IJCNN, but the timeline was too tight. We eventually pivoted to UAI—a harder target, but the only conference with a relevant scope and an open deadline.

That period was brutal. I worked from noon to midnight most days, and the experiments weren’t producing the results I’d hoped for. A lot of effort felt wasted. Weifei helped me refine the direction, and I managed to find some alternative angles worth pursuing, but the stress was immense. I kept reminding myself that this was my first time leading a project—failure was acceptable. My parents said the same, though at the time, none of it really landed.

The Turning Point

At my lowest point, my dad insisted I take a break and spend the day with my family. I didn’t want to—I felt I couldn’t afford to stop—but I also couldn’t focus on anything anymore. So I went.

I still remember that afternoon. We walked along the boardwalk by the sea near our home. I hadn’t been outside in weeks. Shenzhen, the city I grew up in, had changed—new cycling paths, new buildings—and I’d noticed none of it, absorbed as I was in deadlines and ambition.

Standing there, the things I’d been chasing suddenly felt remote. What registered instead was simpler: the light on the water, the grey in my parents’ hair, the quiet fact of being alive and present.

That day clarified something for me. Achieving greatness isn’t the only thing that matters. I want to maintain close relationships, pursue things that genuinely interest me, and become a more complete person. Money is useful, not ultimate. Recognition is nice, but inner peace is worth more. Hard work is important—but it’s a means, not the entire point. Push too hard in the wrong direction, and you break yourself for nothing.

Then I Realized

I came to an honest conclusion: I’m not suited to be a researcher—at least not yet. Research rewards you only after sustained, often painful effort, and it frequently drains the enjoyment from the process itself. I don’t thrive in an environment where the value of what I’m pursuing is constantly uncertain. Some people achieve breakthroughs after thousands of failures. That’s admirable, but it doesn’t mean it’s the right path for everyone, and I’ve come to accept it’s not right for me at this stage.

I don’t say this dismissively, but I believe most academic research has limited real-world impact, and producing truly significant work typically requires a large, well-resourced team. Our group was good—Prof. Jie’s guidance, Weifei’s talent, external consultants—but we were small. In practice, I ended up solving most problems alone, which was more than an average undergraduate should have been expected to handle. I didn’t fail, but I wasn’t exceptional either. Without the team, I wouldn’t have published anything. And in academia, if you’re not rising fast, someone else is, and much of your effort quietly becomes irrelevant. It’s not as grim as that sounds in every case, but that’s how it felt.

What I did feel confident about was my ability as an engineer. I can pick up new technologies quickly and apply them to build things of value—that process genuinely excites me. Engineering still involves solving novel problems, but the approach is different: you’re assembling known tools and extending established methods, not groping in the dark. I want to build something useful that many people can actually use—something elegant in its simplicity.

Research, meanwhile, had been preventing me from developing those engineering skills in a satisfying way. Everything I learned was in service of producing papers, not building real systems. The research itself wasn’t wrong, but I wasn’t ready for it. So I made a difficult decision: this would be my last research project as an undergraduate.

It’s Hard to Say Goodbye

The UAI 2025 deadline arrived on January 12th. After that walk with my parents, I submitted as soon as our team completed the final review. The relief when I clicked “submit” was immediate and enormous. I spent the remaining two weeks of winter break with family and friends, making up for the month I’d lost to the paper. It was the best holiday I’d had since entering university.

After two weeks of rest, the guilt set in. I kept telling myself I should be reading papers, staying productive. Though I no longer enjoyed it, I resumed when the spring semester started. I didn’t miss the work, but letting go of research was harder than I expected—it had become part of my identity, and identities don’t shed easily.

I talked to Weifei about it, but I don’t think he fully understood. He encouraged me to stay in research. Prof. Jie said I was better than most undergraduates at our college and that it would be a waste to quit now. He argued that engineering offers faster early growth, but research has greater long-term potential. Their words made the decision harder. I ultimately declined their advice and left the group, but the doubt lingered.

Shortly after, I joined an engineering-focused lab at BUPT. The work was less demanding than research, but I wasn’t as satisfied as I’d anticipated. I felt adrift—not challenged enough to be engaged, not free enough to be happy.

Spiraling

Anxiety crept in. “Should I go back to research?” I kept circling that question, only to remind myself: “That’s not what you want.” But the more I deliberated, the more confused I became. Eventually, I began fantasizing about a compromise—a group that balanced research and engineering, where I could switch between the two as I pleased. This coincided with summer camp admissions at stronger universities, which seemed like an opportunity. Encouraged by Weifei and others, and seduced by the prospect of a lucrative career as an algorithm researcher, I made a rash declaration: I would apply to labs at top universities. If that didn’t work out, I’d go abroad rather than “settle” at BUPT. In hindsight, it was irrational—I was simultaneously resentful of research and terrified of life without it, propelled by others’ expectations and my own vanity.

But the truth hadn’t changed. I didn’t enjoy research. I tried working with Weifei again, and my performance was worse than ever—I simply lacked the motivation. Yet I’d committed publicly to a plan, and retracting it felt impossible. Stuck in that bind, I left my dorm one evening and rode a shared bike for hours through the city, trying to think my way out. I couldn’t.

Maybe This Is the Answer

I’m generally a rational person, and after a few days, I recognized I’d lost perspective. I called my parents and laid everything out. Talking to them restored my equilibrium.

I had believed that research would elevate me above industry peers, making wealth and status more accessible; that remaining a software engineer was somehow settling; that committing to research would cost me time with the people I care about. None of these held up under scrutiny. Research doesn’t guarantee wealth. Software engineering isn’t a lesser career—it may even be a more demanding and rewarding one. My research accomplishments owed more to the talented people around me than to any innate aptitude. And research, while time-consuming, doesn’t consume everything. I had inflated a straightforward career decision into an existential crisis.

Underneath it all was an unexamined need to be exceptional—a pressure that had hardened into pride. For years, I’d fixated on earning enough to match or surpass my father. But that afternoon by the sea had revealed something truer: what I actually want is peace. Research isn’t a trophy to chase or a trap to avoid. It’s a discipline that requires skills I haven’t yet developed. Maybe I’ll develop them in time—maybe not. Either way, it’s just one option among many.

What matters now isn’t making the definitive choice in a single anxious moment. It’s moving in a direction that feels right, at a pace I can sustain. For now, I’m setting research aside and focusing on development projects—the work I’d been avoiding while paralyzed by indecision. The choice doesn’t need to torment me. What matters is doing work I find meaningful and growing through it. Research isn’t the enemy; I’d consider it again in the right context—a collaborative, low-pressure environment where leaving is an option if it doesn’t work out. I don’t resent research itself, only the strain and emptiness it brought under the wrong conditions. If it doesn’t work next time either, I’ll move on.

There Is No Easy Way

Recently, the engineering work has grown more demanding. The problems are genuinely difficult, and solving them requires sustained effort. But there’s a satisfaction in the process that I didn’t feel during research. I can see how the pieces connect, how the work might become something real and useful. It’s tiring, but I trust that I can work through these problems given enough time and persistence. That confidence—the belief that effort will reliably lead somewhere—is what was missing before. Both paths are hard. But now I know which kind of hard I’m willing to endure.

Choose the Path You Truly Want

When you find yourself at a crossroads, be honest about what you actually want. If you can’t pursue both paths, commit to the one that matters more to you. If you’re unsure what you want, at least identify what you’re unwilling to sacrifice. Neither path is wrong. It’s only a question of what you value—today, next year, or ten years from now.