Chapter 21: Accepted
The email arrived at 6:47 AM on a Tuesday in November, while Dojun was eating his mother’s leftover kimchi fried rice and reviewing Bridge’s learning module code for the third time that morning.
Subject: ISCA 2007 — Notification of Acceptance
Dear Mr. Park,
We are pleased to inform you that your paper, “Dual-Mode Branch Prediction for Bimodal Workloads: A Power-Aware Architecture,” has been accepted for presentation at the 34th Annual International Symposium on Computer Architecture (ISCA), to be held in San Diego, California, June 9-13, 2007.
Your paper was reviewed by three independent referees. The consensus was that your contribution represents a novel and potentially impactful approach to branch prediction design. One reviewer noted: “This is among the most surprising submissions I’ve reviewed — not because of the topic, but because of the maturity and rigor demonstrated by an undergraduate author.”
Dojun read the email three times. Then he put down his chopsticks, walked to the bathroom, looked at himself in the mirror, and said: “You just got accepted to ISCA.”
His twenty-year-old face stared back at him, unremarkable, slightly puffy from insufficient sleep, with a grain of rice stuck to his chin.
“At twenty,” he added.
The face in the mirror didn’t look like someone who had just received the most prestigious academic acceptance in computer architecture. It looked like someone who needed to wash his rice bowl and get to class.
He washed the bowl, wiped the rice off his chin, and called Kim Taesik.
“I know,” Kim said, before Dojun could speak. “I got the notification ten minutes ago. I’ve been sitting at my desk trying to decide whether to be proud or terrified.”
“Which one won?”
“Both. Simultaneously. It’s an unusual emotional state.” A pause. “ISCA, Park. The youngest Korean author in the conference’s history. The youngest author from any Asian university in at least fifteen years. This is—” Another pause, longer. “This is significant.”
“I know.”
“No, you don’t. You think this is about a paper and a conference. It’s not. An ISCA acceptance at your age turns you from a promising student into a national academic story. The media will cover it. The ministry will notice. Other universities will try to recruit you. And every professor who has been quietly wondering about you will now have a very loud reason to ask questions.”
“You’re saying the attention gets worse.”
“I’m saying the attention changes category. From curiosity to scrutiny. There’s a difference.” He paused. “Come to my office. We need to plan the announcement carefully.”
“When?”
“Now. Bring coffee. The good kind, not mine.”
Dojun bought two Americanos from the campus cafe and walked to the engineering building. The November morning was cold—Seoul’s autumn had given way to early winter almost overnight, and the ginkgo trees that had been gold last week were now bare, their leaves piled in drifts along the walkways like discarded treasure.
Kim Taesik was at his desk, the ISCA acceptance email printed out and placed next to Dojun’s original paper manuscript—the one covered in red annotations from their first round of revisions. He accepted the coffee with a nod.
“The announcement strategy,” Kim said, moving into planning mode with the brisk efficiency of a general briefing troops. “I’ll notify the department chair and the dean today. They’ll want to make a press release—the university loves good publicity, and an ISCA acceptance from an undergraduate is the kind of publicity that gets them government funding.”
“Can we control the narrative?”
“To an extent. I’ll draft the press release myself—emphasizing the research program, the accelerated track, the mentorship. The story isn’t ‘mysterious student publishes impossible paper.’ The story is ‘SNU’s investment in gifted education produces world-class research.'”
“That’s the university’s story. What about mine?”
“Yours is simpler. You’re a dedicated student who worked hard under excellent supervision. That’s it. No mystery, no drama, no quotes about reading MIT papers or ‘just knowing things.’ When reporters ask—and they will—you defer to me. I speak for the program. You speak for the research.”
“Understood.”
“And the conference itself. San Diego, June. You’ll need to present in person. That means international travel, which means a passport, which means—” He looked at Dojun. “Have you ever left Korea?”
“No.” In this life, that was true. In his previous life, he had lived in San Francisco for eight years and traveled to forty-three countries. But his passport was blank.
“Then we need to start the visa process now. The American consulate takes weeks for student visas. I’ll write a letter of support.”
“I can handle the paperwork.”
“I know you can. But I want the university’s name on the application, not just yours. An institutional backing makes the visa easier.” He sipped the coffee. “Good Americano. You should buy me coffee more often.”
“I’m on a student budget.”
“You’re the CEO of a funded startup. Your budget is relative.” The ghost-smile. “Congratulations, Park. You earned this.”
“We earned it. You supervised every revision.”
“I supervised. You wrote. The ideas are yours—the dual-mode predictor, the bimodal distribution analysis, the power optimization. I just made sure you didn’t express them in language that would get rejected.” He stood up and extended his hand. “Shake my hand. This is a formal moment.”
Dojun stood and shook his professor’s hand. Kim’s grip was firm—firmer than usual—and he held it for a beat longer than protocol required.
“I said once that you might be important,” Kim said. “I’m amending that. You are important. Don’t waste it.”
“I won’t.”
“I know.” He released the handshake. “Now go tell your team. And your mother. In that order, because if your mother finds out from a newspaper before she hears it from you, I will not be able to protect you from the consequences.”
Hana screamed.
Not a polite, contained exclamation. A full-throated, office-shaking scream that rattled the whiteboard markers in their tray and brought the biotech startup next door to their door with concerned expressions.
“Everything okay in there?” the biotech founder asked, peering through the glass.
“We just got accepted to the biggest computer architecture conference in the world!” Hana yelled.
“Is that… good?”
“It’s INCREDIBLE! Go away!”
The biotech founder retreated. Hana turned back to Dojun, eyes blazing.
“ISCA. Park Dojun. ISCA! In San Diego! In America!” She grabbed his shoulders. “Do you understand what this means for Bridge? We have a co-founder who’s presenting at ISCA. That’s not a student startup anymore—that’s a research-backed technology company with genuine academic credibility.”
“It’s a paper about branch prediction. It’s not directly related to Bridge.”
“It doesn’t matter. Credibility transfers. When Choi Eunji tells other investors about Bridge, she can now say ‘the CTO published at ISCA.’ That sentence opens doors that no demo can open.” She was pacing, her mind racing ahead. “We need to update our pitch materials. And the website—Bridge needs a website, Dojun. A real one. With a team page that lists your ISCA publication.”
“We don’t have a website.”
“Exactly! That’s insane! We have five hundred beta users and no website! I’m designing one tonight.” She grabbed her sketchbook and started wireframing, standing up, pen flying. “Clean, minimal, professional. Team page, product description, beta signup—wait, the beta is full. Waitlist. We need a waitlist.”
“Hana. Breathe.”
“I’ll breathe when the website is live.” But she paused long enough to smile at him—a smile so bright it could have powered the office’s intermittent WiFi. “I’m proud of you. Genuinely, deeply proud. You deserve this.”
“The team deserves—”
“No. Stop. You wrote that paper. You ran those simulations. You stood in front of sixty professors and defended your thesis. The team built Bridge together, but ISCA is yours.” She stepped closer. “Take the compliment, Dojun. You’re getting better at many things. Taking compliments is still a work in progress.”
“…Accepted.”
“See? Progress.” She kissed his cheek—the same spot as the subway platform, brief and warm. “Now call your mother before she finds out from the news.”
His mother answered on the first ring, as always.
“Dojun-ah? Why are you calling in the morning? Are you sick?”
“I’m not sick, Mom. I have news.”
“Good news or bad news?”
“Good news. My research paper was accepted at an international conference. In America. San Diego.”
Silence. He could hear the market in the background—the familiar orchestra of vendors and customers and commerce.
“America?” she said. “You’re going to America?”
“In June. For five days. To present my research.”
“Your research on the computer chips?”
“On processor design, yes. It’s the biggest conference in the field. Being accepted is—it’s a very big deal, Mom.”
“Of course it’s a big deal. My son is going to America to talk about computers.” She said it the way she would say “my son cured cancer”—with absolute, unquestioning pride, regardless of whether she understood the details. “Wait. How much does it cost? The airplane—the hotel—”
“The university pays for it. Fully funded.”
“The university pays? For you to go to America?”
“Because my paper was accepted. It’s a university-funded research trip.”
“Aigoo, Dojun-ah.” Her voice had gone thick. “Your grandmother—she always said, ‘That boy will go somewhere.’ She didn’t mean literally. She meant in life. But you’re doing both.”
“I’ll call you from San Diego.”
“You’d better. And eat there. American food is strange—everything is too big and too sweet. Bring snacks from Korea. And take kimchi. You can freeze it and—”
“You can’t bring kimchi through customs, Mom.”
“Why not? It’s food! Everyone needs food!”
“American customs has different opinions about fermented cabbage.”
“American customs has bad taste.” She sniffed. “Come Saturday. I’ll make special japchae. The celebration kind, with beef and extra vegetables.”
“I’ll be there.”
“And Dojun-ah?”
“Yeah?”
“I’m proud of you. I don’t understand what you do. I probably never will. But I know it’s good because good people keep telling me it’s good. Your professor called me, you know.”
“Kim Taesik called you?”
“Last week. He said, ‘Mrs. Park, your son is going to be famous.’ I said, ‘He already is—he’s famous in this market for being the skinny boy who doesn’t eat.'” She laughed. “He laughed too. He has a good laugh, your professor. Even if his coffee is terrible.”
“You know about his coffee?”
“He told me. He said it’s legendary. I told him legendary is not the same as good. He agreed.”
Dojun laughed until his eyes watered. The image of Kim Taesik and his mother bonding over bad coffee was so unexpected and so perfect that it felt like a scene from a drama he would have watched as an old man in a hospital bed.
“I love you, Mom.”
“I love you too. Eat breakfast.”
That evening, Seokho called.
“ISCA,” he said, without greeting. “I saw the notification on the ISCA website. Your paper is listed. ‘Park Dojun, Seoul National University.'”
“News travels fast.”
“I monitor ISCA acceptances. It’s the most important conference in architecture. Knowing who’s presenting there tells me where the field is moving.” A pause. “You’re presenting there. At twenty. That’s—” Another pause, and Dojun realized he was hearing something unprecedented: Seokho struggling for words. “That’s beyond what I expected, even from you.”
“High praise from the man who expects everything.”
“I expect everything from myself. From you, I’ve stopped expecting and started observing. Every time I think I’ve calibrated my model of Park Dojun, you exceed it by an order of magnitude.” He was quiet for a moment. “I have a proposal.”
“I’m listening.”
“I’m presenting a paper at SIGCOMM next year. Network consensus protocols. The conference is in Pisa, Italy, August 2007.” He paused. “I want to collaborate on a follow-up paper. Your dual-mode prediction architecture, applied to network traffic classification instead of processor branches. I think the bimodal distribution model generalizes—network traffic is also bimodal, with predictable flows and unpredictable flows.”
Dojun’s mind raced. Seokho was right—the bimodal model did generalize. In his previous life, a similar cross-domain application had been published in 2019 by a team at MIT, and it had become one of the most-cited networking papers of the decade.
But building that paper with Seokho, in 2007, as co-authors—that was a collaboration that had never happened in the original timeline. Dojun and Seokho had been competitors, not collaborators. They had published competing papers, filed competing patents, built competing companies. The idea that they could work together was so foreign to the original timeline that it felt like a violation of physics.
And yet it was the obvious right move. Seokho had the networking expertise. Dojun had the prediction architecture. Together, they could produce something neither could build alone.
“I’m interested,” Dojun said. “But I need to understand the scope. A SIGCOMM paper is a major commitment. I’m running Bridge, taking graduate courses, and presenting at ISCA. My bandwidth is—”
“Limited. I know. The proposal is this: I lead the networking side—experimental design, simulation setup, data collection. You contribute the prediction model adaptation and the theoretical framework. We meet biweekly, alternating between Seoul and Daejeon, to review progress. Target submission: April 2007.”
“That’s ambitious.”
“Everything worthwhile is ambitious.” A beat. “Park. This isn’t just about a paper. You know that.”
“What is it about?”
“It’s about whether the two best computer scientists of our generation—and I use that phrase with deliberate precision—can build something together instead of building against each other. Every other outcome of our relationship is predictable: rivalry, competition, parallel paths that never converge. Collaboration is the only interesting option.”
Dojun closed his eyes. In his first life, the rivalry with Seokho had consumed decades. Patent wars, talent poaching, public feuds disguised as technical debates. It had made both of them sharper—and both of them lonelier. The only time they had been honest with each other was at that bar in Itaewon, at fifty, after too much soju, when Seokho had said: “We wasted thirty years competing. Imagine what we could have built together.”
Now, at twenty-one, Seokho was saying the same thing. Forty years earlier. Without the soju.
“Let’s do it,” Dojun said.
“You didn’t hesitate.”
“I’ve learned that the best decisions are the ones you don’t need to think about.”
“Interesting philosophy. Dangerously impulsive for an engineer, but interesting.” The rare Seokho almost-smile was audible through the phone. “I’ll send you the preliminary framework by Friday. We start the week after.”
“And Seokho?”
“Yes?”
“I appreciate it. For proposing this. It means more than you know.”
“It means exactly as much as I intend it to. I don’t do things accidentally.” A pause. “Take care, Park. And congratulations on ISCA. You’ve earned the right to be terrified of presenting in San Diego.”
“I’m not terrified.”
“You should be. The Q&A at ISCA makes Kim Taesik’s colloquium look like a kindergarten circle time. But you’ll survive. You always do.”
He hung up. Dojun sat in his apartment, phone in hand, staring at the water-stained ceiling. The lopsided heart stain was still there—unchanged, permanent, a fixture of a life that was changing in every other way.
ISCA in June. SIGCOMM collaboration with Seokho. Bridge’s beta growing. Hana designing a website. His mother making celebration japchae. Kim Taesik calling her to tell her the son she had raised alone was going to be famous.
The web of connections was growing denser, stronger, more real. Each thread—professional, personal, familial—reinforced the others. Pull one, and the whole structure held. Break one, and the others compensated.
This was what he had failed to build the first time. Not a company or a reputation or a fortune. A web. A network of people who showed up for each other, who held each other’s weight, who didn’t let go when the wind blew.
He opened his journal and wrote a single line:
November 2006. Eight months since I woke up. The web holds.
Then he closed the journal, opened his laptop, and began adapting the dual-mode prediction model for network traffic classification.
The cursor blinked. The code flowed. And outside, Seoul settled into its November rhythm—cold, clear, the kind of winter stillness that made every sound sharper and every light brighter.
Eight months. One paper accepted. One company funded. One rivalry transformed. One relationship growing.
And seventeen more chapters of the first volume of his second life, still waiting to be written.
He typed. The web held. The future, for the first time, felt like something worth building—not because he knew what was coming, but because the people building it with him made the uncertainty bearable.