Skip to main content
  • Home
  • Library
  • Read
    • Popular
    • Charts
    • Browse
    • Categories
    • Wikis
    • Glossary
  • Request
  • Activity
  • Leaderboard
  • Help
Our Apps
  • DaoSearch
  • Narratorr
Social
  • Discord
  • Patreon
The AI Wave Tycoon's Rise
Chapter 493

I Don't Understand Games, but I Understand Products

1,656 words 6 min read 1 view 14d ago Source ​
Skip to the end of the chapter
Prev
Next

Banishing the crazy idea that suddenly popped into his mind, Han Luyi had to admit that catching up with others' decades of accumulation in just a few years was a difficult path, especially considering that it required the perfect combination of favorable timing, location, and people.

But the more difficult the journey, the more worthwhile it is.

The concept of global division of labor may not be wrong, but when cutting-edge technology is controlled by others, all development is built on castles in the air, and such a division of labor seems ridiculous.

Origin has made great progress by leveraging Han Luyi's breakthroughs at the model level to bring together the strength of domestic manufacturers that were previously operating independently through open source.

But after that, there's still a long way to go in terms of mass production, improving yield, reducing costs, and even overcoming limitations at the production equipment level.

"Computing Power has always been a key focus for us, but we can't always rely on cost-cutting measures," Han Luyi's thought. "Right now, my focus is mainly on two aspects: firstly, before the L200 can begin mass production, we need to acquire as many cards as possible from various domestic Computing Card manufacturers, and secondly, we need to improve efficiency through a unified architecture to obtain more effective Computing Power."

"As for the other aspect, it depends on the training progress of the new model. The Computing Power for pre-training Glutinous Rice Ball 2.0 is currently unavailable. I asked Wenyuan to Model Training first. When he reported to me this morning, he said that the base has been running successfully. The next step is to fine-tune the model. I can hear your opinion on which direction to work on first."

Origin current structure, apart from functional departments such as administration, human resources, and legal affairs, can be mainly divided into three sections: the Beijing Research Institute led by Zhao Wenyuan is mainly responsible for model development; the Computing Power business unit led by Jiang Songran is responsible for solving Computing Power supply; and the comprehensive productization section led by Su Niannian is responsible for selling the capabilities of the model in various ways.

Of these three sectors, the Research Institute and Computing Power sectors are the biggest money-eaters, while only the sector led by Su Niannian is profitable.

Although she never said it out loud, Su Niannian was clearly under a lot of pressure in this regard.

Therefore, faced with the unexpected good news, Su Niannian's deliberately stern face relaxed, and she quickly got into work mode.

She asked, "How much performance will the Lightweight Model sacrifice? How much cost will it save? What are the fine-tuning options?" Han Luyi chuckled at Su Niannian's three questions in quick succession, "The cost will be lower, and the performance degradation in specialization should be minimal. However, for specific data, your product department needs to directly coordinate with Wenyuan. Feel free to raise any needs you may have. Until I free up the Computing Power for training 2.0, I'll have them prioritize meeting your requirements for the Lightweight Model."

Su Niannian realized she had asked too hastily, paused for a moment, and said, "Then thank you Boss Han, for your support."

Then she thought for a moment and said, "Now it seems that there are two directions that are most worth investing in. One is Riding the Wind. Xiao Jiang mentioned to me before that after the Douzhi Model is released, it can be used in Riding the Wind for Task Scheduling, routing planning tasks to the flagship model and assigning execution tasks to the Lightweight Model. Originally, we planned to call a third-party Lightweight Model, but if our own Lightweight Model is released, it can be used here, and the demand for Computing Power can be further alleviated."

After Su Niannian finished speaking, she glanced at the tablet on the table and continued, "Another aspect is games."

Han Luyi raised an eyebrow and asked, "It only went live yesterday, and you've already decided to raise its priority?"

"The budget hasn't been adjusted yet, but I've carefully reviewed the data Boss Zhang brought, especially comparing it with publicly available data from industry survey reports. It's clear that our attempt to integrate Large Models into game scenarios is very promising. We've already achieved such results in just a test event for a relatively unpopular older game. If we invest more resources, I think this could be a key area for development."

Su Niannian spoke earnestly, leaning slightly forward. A wisp of hair perked up on her head, swaying gently with Su Niannian movements, probably unnoticed even by herself.

Han Luyi looked at that tuft of hair and couldn't help but tease, "Didn't you say before that you'd never made games before? How come you're so into it now?"

Su Niannian couldn't help but roll her eyes at him, then immediately realized that what she was doing was inappropriate, and explained, "I haven't made games, but I have made products. Data doesn't lie. Looking at it from a broader perspective, the conversion rate between game time and revenue is basically proportional. In other words, our current direction is at the forefront of the entire game industry. Combining the Large Model with the game experience could be a huge market, and it's worth our investment."

"Originally, your investment in games was considered in terms of data platforms, but judging from market feedback, the business opportunities here may be even greater than we initially anticipated. Didn't you previously mention considering the collaboration between the gaming industry and Brain-Computer Interfaces? Brain-computer interfaces may still be too far off, but the collaboration between models and games is already feasible. My main concern before was the cost issue, but if this problem can be solved by Lightweight Models, that concern will disappear."

After saying that, she added, "I was just about to report this, but you interrupted me. The effect of this event is gradually going viral. Several game companies that I contacted before have sent messages to inquire about how we did it and whether they have any plans to license the models."

Han Luyi asked with interest, "So what was your reply?" "Naturally, nothing. Aside from exploring our own game, we don't have the extra Computing Power to collaborate with other companies right now."

Han Luyi glanced down at the tablet on the table, pondered for a moment, and said, "It seems that the prospects for developing in the gaming industry are indeed greater than I initially thought. Besides providing us with data, the gaming industry can really make money."

Su Niannian smiled and said, "Don't worry about making money, Boss Han. Just leave it to me to figure out how to maximize the existing resources."

The two exchanged a glance, then looked away.

Han Luyi coughed lightly and said, "I'll leave the product development to you. Ideally, you should establish a social data collection pipeline through games within three to six months. You'll have to figure out how to balance the resources, but one thing is certain: the majority of Computing Power should still be allocated to Wenyuan first; research and development should be the priority."

Su Niannian nodded.

"By the way, about the software joint venture with Icecore that I mentioned before, I've already talked to Liang Wei and Li Wen, and they've both agreed. Old Chen and Xiao Jiang are your direct subordinates, so could you go and talk to them for me?" Han Luyi said.

Su Niannian looked up at Han Luyi and blinked.

While job transfers are normal, simply removing people from their jobs leaves them to coordinate their own work, creating unnecessary extra workload for Old Chen and Xiao Jiang.

Are you asking me to do something that offends people?

"What if they ask me for them again?" Su Niannian asked.

"The company is expanding rapidly, there's nothing we can do about it. We'll promote some while also looking for suitable candidates outside," Han Luyi quickly changed the subject. "Oh, and one more thing, my brother just finished his college entrance exams, and my mom's planning to get him an internship here. Do you have any suitable positions?" Su Niannian's first reaction was, "You have a younger brother? How come I've never heard you mention him before?"

"My cousin is ten years younger than me, but I still feel like he's a little kid in open-crotch pants."

Su Niannian seemed relieved after hearing this. After thinking for a moment, she said, "Your cousin just finished the college entrance exam. Is he really here to learn something, or is he just trying it out? In any case, he probably doesn't want others to know his identity, right? This batch of students who came for summer internships is already halfway done, and the positions in each group have been assigned. If we squeeze someone in now, anyone would find it strange."

Han Luyi, "It's not even certain whether he'll come or not; I haven't even asked him what he thinks yet."

That's true, though. If Zhou Muzhi were to come to the company for an internship and use Han Luyi's name to bully everyone, it would be pointless.

Han Luyi flash of inspiration and said, "How about we let him be Xiao Jiang's assistant? I saw last time that she was quite envious of Liu Yu. Hmm, it's not settled yet, you can mention it to her while you're at it."

Letting kids take care of kids—what a brilliant idea!

End of Chapter
Prev
NovelSummaries
Next

Explore the wiki

Han Luyi
Su Niannian
Zhao Wenyuan

Comments

No comments yet.
AboutStatsAPIChangelogStatusContactPrivacyTermsDMCA
© 2026 RandomTranslator·v3.26·RSS