CHAPTER TWELVE
The Leaderboard
I stayed up until 2 a.m. reading papers about animal behavior and domestication. A poor decision, but I was desperate. Then I had a horrible dream about wolves attacking the humans that tried to domesticate them. I woke up in a tangle of sheets to the persistent bleat and blinding glow of my multialarm. The volume and brightness told me I’d overslept. Shit.
It was 7:45, not as bad as I thought. The jalopy fired up on my second try. I slurped hot coffee on my way in. Scalded my tongue something fierce, but it got my neurons firing. By the time I hit my workstation, I had four new ideas I wanted to try to get that aggression score down. If the field hadn’t changed much since yesterday, I might even catch up.
My naive optimism lasted right up until I logged in. E-mail alerts flooded the screen.
Korrapati: 69.20.
Wong: 76.89
Wong: 75.01
Wong: 73.93
Wong: 72.55
Wong: 70.48
Wong: 69.12
O’Connell: 73.11
Well, crap. So Korrapati stayed late, O’Connell came early, and Wong never even left. Jesus. I rolled back my chair and leaned over the divider. Wong slouched behind his wall of empty drink containers, with no less than four projection monitors hovering in front of him.
“Hey Wong, busy night?”
He swiveled enough to grin at me. “Very busy!”
“You ever sleep?”
He shook his head. “Sleep when we domesticate the dragon.”
I chuckled and returned to my desk. That’s the Shenzhen work ethic, all right. I couldn’t argue with the results, though; according to the latest simulation run, Wong had the lead. I pulled up my own prototype and started modifying the mitochondrial pathways. Mitochondria produce energy for cells, and thus affect numerous processes, from stress response to memory to overall longevity. The little power plants even have their own genetic code, a tiny circular chromosome with a couple hundred genes. Most engineers don’t even touch the mitochondrial genome, because it already works quite well. Mutations tend to cause subtle degenerative diseases.
But it occurred to me, in my sleep-deprived morning state, that the speed and ferocity of Design 48’s attack had required a lot of mitochondrial energy. Maybe tamping down the energy production wasn’t a bad idea. Nature had done some of the hard work for me already—we already knew dozens of mutations that altered energy production and overall function of the mitochondria. I selected a couple of these, put one back, chose another. Like a shopper at a genetic engineering supermarket.
Parker: 82.18
Well, I still hadn’t broken 80 but that marked a big improvement. Even better, I doubted any of the other designers thought to try this route. The mitochondrial genome is so tiny, it’s easy to forget.
I turned my attention to brain development. So many of the critical neuronal connections form well before an animal is born, so it stood to reason that with some rewiring early on, we might get a better handle on temperament. But unlike the mitochondrial changes, I couldn’t go browsing the natural catalogue of brain development mutations. They tend to be lethal. That simple, unavoidable fact hinted that perhaps I shouldn’t be tampering with such a crucial element of dragon biology, but . . .
Korrapati: 58.63.
Hell, I had nothing to lose.
I plunged into the dark recesses of developmental neurobiology. The genetic program that controlled this process was like an elegant orchestra—genes turning on and off in rhythmic fashion, new patterns rising and falling as the vital connections formed. I harnessed those elements and rewrote their playbook. Well, maybe that’s an overstatement. I nudged their genetic fate a bit here and there. I forged new links between sensory perception, memory, emotional response.
My free points plummeted as DragonDraft3D recomputed the intelligence quotient. The dragon seemed smarter, though in fact what I’d hoped to grant it was awareness. Awareness of itself, its surroundings, and its human owners. A reptile that relied on intellect more than instinct might see the long-term benefits of human partnership. Time for the real test. I sent my new prototype to the simulator.
Parker: 62.61
“Holy crap,” I breathed.
No one snorted with derision this time. The air hung heavy with tense concentration. I knew they had to be thinking, How did he do that?
I smiled and got right back to it. Maybe it was my imagination, but DragonDraft3D felt easier to navigate. I’d been so focused on the tough developmental changes that I hadn’t noticed it happening. The further I climbed up the learning curve, the more naturally my next moves came to me. At first, Evelyn’s interface design felt almost random. Now I saw this higher-order structure to it: physical to intellectual to behavioral capabilities. I could probably design my current prototype from scratch in under an hour.
Those days blurred together into an endless block of caffeine-fueled design. The competition consumed every part of my mind, to the point where I forgot to consider that winning this contest was just a means to an end. The deadline crept ever closer, and I began to close the gap on the rest of the team.
Then we hit the fifty-point plateau.
Korrapati got there first. Her prototype hit 51.33, then 50.08. On her next run, it bounced back up to 50.73. I didn’t notice this until later—I’d forced myself to stop checking the scores the moment they came in. I was still in the low sixties. My competitors’ successes only harmed my productivity, and I couldn’t afford to fall into a slump.
“Noah?”
I recognized Korrapati’s voice, but it took a moment to register the fact that she’d spoken to me. Everyone in the design lab had pretty much stopped talking to each other as we hit the home stretch of the competition. I leaned back. “Priti. Hey.”
“Have any plans for lunch? The food trucks are here today.”
I nudged my insulated lunchbox farther under my desk. “That sounds great.”
Should I invite Wong? He hadn’t eaten, or even moved from his chair. But he’d insist on riding his scooter, which would take forever. Besides, I kind of wanted to hear what she had to say. So I left him and followed her to the elevator. A couple of dungaree-clad wranglers occupied the one that stopped for us. We rode it down in silence. That silence pressed on us as we made our way past the desk.
“Hey, Virginia,” I said.
She brightened. “Hi, Noah!”
Korrapati did a double-take. Inwardly, I smiled.
A parade of hybrid-fuel trailers lined the curb on the street outside Reptilian’s front door. The bright, garish paint jobs assaulted my eyes like an attack of mismatched tropical birds. The lines in front of most of them were three deep.
“What do you feel like?” Korrapati asked.
“Ooh, how about Asian fusion?”
“Are you suggesting that because I’m Asian?”
Oh, hell. “No, I, uh—” I stuttered.
She flashed me a perfectly wicked smile. “Just messing with you.”
I scowled at her, mock-serious. “You are vicious.”
“Only when I want to be.”
“Shouldn’t the person at the top of the leaderboard be the nicest right now?”
She schooled her face back to neutral quicker than I’d have thought possible. “That’s only temporary.”
“Seems to me you’re the one to beat.”
She stopped short at a lime green food truck. “Well, here’s your Asian fusion.”
Asian-Mexican fusion, I didn’t dare say. I ordered something that held a prominent place at the top of the truck’s digital menu. My order arrived in a cardboard box in about two minutes. The cardboard didn’t quite burn my hands, but it weighed over a pound.
We sat on the thick concrete curb that doubled as a security barrier for Reptilian’s glass entrance.
“What did you order?” she asked.
“Something called carnitas fries.” I flipped open the box lid. The aroma set my mouth to watering. “Sure smells delicious.”
“You chose well. That’s their specialty.”
“What did you get?”
“Grilled cauliflower.”
I tried and failed to conceal my reaction.
“What?” she laughed. “It’s good.”
“Oh, I’m sure.” If you’re a rabbit.
We ate in silence for a moment. The carnitas fries were still piping hot, but damn good. I couldn’t identify half of the toppings that weren’t meat, but I ate it all anyway. We made small talk, while I waited for her to tell me what this lunch was really about.
“How do you like working at Reptilian?” she asked.
“It’s good. Challenging.” I paused. “And a little different than I expected.”
“How so?”
I shrugged. “I thought we’d be churning out a lot more dragons, for one thing.”
“We used to,” she said, with an almost wistful air. “That’s all I did, when I first started here.”
“And now?”
“I suppose the demand tapered off. For wild ones, at least.”
I feigned seriousness. “So basically, what I’m hearing is that you hogged all the dragons.”
She gave a little gasp but smiled. “I make no apologies.”
“Well, if we can crack domestication, there should be plenty of work to go around.”
Her grin faded, and the sharp lines found their way back to her face. “Yes. Domestication.”
There’s the opening she wanted. I braced myself, because I half-expected she was about to tell me that I had no chance at winning this competition.
“I’d like to learn more about how your simulator scores aggressiveness,” she said.
“I posted the code. You should be able to read it.”
“Yes, I could read ten thousand lines of code. Or I could simply ask you.”
I chuckled. “Good point. Well, it’s a mixture model with two primary dimensions: the chemicals in the brain—”
“Neurotransmitters?”
“And hormonal regulators, yes. The other dimension tries to capture psychological intent. You know, what drives the animal.”
She nodded along as I spoke, as if unsurprised.
She probably did read the code but wanted me to confirm it.
“Is there any sort of a manual threshold? A minimum value they won’t go below?”
I smiled. “At around fifty, you mean?”
“Yes.”
She’s stuck. The realization gave me a glimmer of hope, because if Korrapati couldn’t improve, maybe I could catch her. Then again, it didn’t seem fair to hold up the best designer in the group out of pure selfishness. “It sounds like you’ve found the best answer, but for only one of the two dimensions. I’m guessing the first one?”
Her eyes widened enough to confirm it.
“Yeah. Intent is a much harder nut to crack,” I said. “There’s so much latent instinct buried in the dragon genome. Stuff we probably don’t understand at a deeper level.”
“Thousands of years of evolutionary pressure, much of it subtle.”
“Exactly.”
“Thank you for the insight. You didn’t have to help.”
“No problem. This contest is just for fun, right? We should be helping each other.”
Furrows appeared in her brow. She stopped eating. “Have you ever heard the joke about the campers and the grizzly bear?”
“I don’t think so.”
“These campers encounter a grizzly bear with her cubs, so they take off running. One of them asks, do you think we can outrun her? And the other one calls back, It doesn’t matter if I can outrun her. I just have to outrun you.”
I chuckled. “Okay, I have heard that one.”
“It’s kind of similar to the leaderboard, don’t you think?” she asked.
“How so?”
“We all want to win the contest. But given the company’s financial crunch, the most important thing is not to come in last.”
“Why?” I felt like coming in last was almost a guarantee for me at this point.
“The worst performer might not have a job when this is over.”
I felt like she’d punched me in the gut. “Do you think Evelyn would do that?”
“She might not have a choice.”
“Well, that kind of sucks.” It made sense, though. Evelyn needed designers who could meet tough challenges. And hell, I’d given her a way to score us, to make our performances quantifiable. I’m such a moron.
Korrapati sighed and squared her shoulders. “Well, I guess I’d better get digging. Thank you for the help.”
“Sure. But I think it’s only fair that I share my ideas with the rest of the team.”
She pressed her lips together but nodded. “That’s your choice.”
When I got back to the lab—feeling nervous, guilty, but oddly satisfied from the carnitas fries—I did just that. Sent an e-mail to the whole design team about the nature of the aggression score and my best guess on the cause of the fifty-point plateau. Of course, none of them believed me. They pushed ahead with their designs, figuring they could get around it.
Wong: 50.34
O’Connell: 50.28
O’Connell: 50.42
Wong: 50.83
Wong: 50.65
Wong: 50.18
Wong: 50.30
“Give it up, Wong,” I called over the divider.
He rolled out far enough to look at me. “No giving up.”
“You’re not in Shenzhen anymore.”
A cloud passed over his face. “Without this job, I go back.”
“Why wouldn’t you want to stay?”
He shrugged. “H-1 visa rules. If I lose job, or company closes, I must return.”
“Reptilian’s not going to close.”
“I would like to have your confidence.”
“Come on, Wong. This company makes dragons for the world. It’s invincible.”
He turned back around to his monitors, shaking his head. “No company is invincible.”
I’m not sure why Wong’s offhand comment rattled me, but it did. If Reptilian folded, its assets would be sold off piecemeal. The God Machine first of all. Then I’d never have a chance to use those to model Connor’s mutation. Now that I thought about it, achieving dragon domestication seemed simple by comparison. There has to be a way to crack it.
And by God, I’d find it or die trying.