Podcasts · Video · Long-form
You set the question. I’ll bring first-hand cases, public evidence, and a clear point of view. My work has spanned economics, data, and AI at Amazon, Meta, Tencent, and Statsig; today I build courses, community, enterprise programs, and long-form media through Superlinear Academy.
The idea behind that work is MAKE WHAT LASTS. It keeps leading me back to the same tensions: what becomes more valuable as execution gets cheaper, how organizations learn whether a judgment is right, and what turns a first result into lasting work.
Cornell Economics PhD
Amazon · Meta · Tencent · Statsig (acquired by OpenAI)
Superlinear Academy · 20,000+ community members
YouTube · Bilibili · Xiaohongshu · 400K+ followers
WHERE AN EPISODE COULD BEGIN
There is no need to fit my entire background into one episode. Pick one tension your audience genuinely faces; we can follow it through mechanisms, first-hand cases, and consequences.
From ‘stop using ChatGPT’ to agentic work: what becomes valuable when execution gets cheaper, and why judgment, craft, and staying power return to the center.
AI Builders · creator work · real member projects
Across Amazon, Meta, Tencent, and Statsig: how data, experiments, incentives, and power can bring an organization closer to reality—or help it rationalize a mistake more elegantly.
Economics · data science · big tech and startups
Return to what was actually public before GPT-4, separate understanding a mechanism from guessing an outcome, and inspect the calls that later needed correction.
A dated public record from 2021 and February 2023
What people are really paying for in education: structure, judgment, feedback, trust, and a peer environment—and how to preserve the quality of that environment as it grows.
AI Builders · Superlinear Academy · courses, community, and membership
Other grounded directions include experimentation, creator economics, China–U.S. technology, and how AI education and community should be built. One real question is usually stronger than a complete tour.
WHERE THE VIEW COMES FROM
My work has moved from economics and data science into product decisions, organizational leadership, startup execution, and AI education. That makes it possible to connect what a model can do with what a company will adopt—and what changes for the person doing the work.
Cornell · Amazon · Meta · Tencent · Statsig
In 2021 I described the ceiling of the dominant AI paradigm. In February 2023, before GPT-4, I argued that ChatGPT had begun to break it—and reasoned forward to cheaper inference, tool use, persistent memory, and AI-native systems. The original probabilities and later corrections remain public.
A dated, public chain of reasoning
At Superlinear, the ideas have to survive contact with a system I am building: a free community, AI Builders, a year-round membership, enterprise programs, and more than 200 public conversations with researchers, founders, and operators.
Superlinear Academy · AI Builders · enterprise AI programs
GUEST APPEARANCES
Six guest appearances are selected below. Outside media, I also work on AI training and programs with teams at Tencent, Xiaohongshu, Meituan, and DoorDash; the complete 23-episode archive remains at the end of the page.
The Build Log
#9 Is power at work an illusion? Borrowed authority and real influence
Grad Lounge 不止学术
118. From the ivory tower to real capability: building personal value in the AI era
INDIGO TALK
Quit ChatGPT to become an AI orchestrator / AI as a second Renaissance · EP46
硅谷101
E195 · From tool to partner: seven experienced AI agent users
商业 WHY 酱
S3E03 · Before asking how to profit, ask the basic questions about ChatGPT
What’s Next｜科技早知道
How should we respond to ChatGPT? Highlights from a closed-door market discussion
PROGRAM INVITATIONS
Send the show or channel, the question you most want to pursue, and a rough format and timing. If I have something genuinely new to add, we can set a time and prepare around the show.
Discuss an episodeBusiness contact: Miao (喵老师) | WeChat: FM13870617
TWO COLLABORATIONS, TWO DIFFERENT TESTS
Koji’s team produced and published the first collaboration independently. The Tulong collaboration used both teams, two edits, and several channels. Together they answer different questions: whether the conversation holds on its own, and what can happen when both sides build around it.
CASE 01 · PARTNER-PRODUCED AND PUBLISHED
Crossing with Koji × Yuzheng Sun
67K
Xiaoyuzhou plays
Show Top 212K
Douyin likes
Account all-time high8,343
Xiaohongshu saves
Account annual high6,502
WeChat Channels shares
Cross-platform sharingThe episode reached 67K Xiaoyuzhou plays. Its clips became Koji’s all-time most-liked Douyin post and annual Xiaohongshu save high. The episode and every clip were produced and published independently by Koji, without co-publishing from my accounts.
Also: 132 Xiaoyuzhou comments; 4,709 Xiaohongshu likes; 8,582 Douyin saves and 1,476 shares.
“A genuinely useful episode. I’m moving from ChatGPT to Codex and building up my context.” — Xiaoyuzhou listenerListen on Xiaoyuzhou
CASE 02 · CO-PRODUCED AND CO-PUBLISHED
Yuzheng × Yang Ying (Tulong)
58.7K
Xiaohongshu likes
Co-published59.6K
Xiaohongshu saves
More saves than likes110K+
YouTube views
New viewers discovered Tulong108K+
Bilibili views
Long-form reachTulong’s team made one cut in the voice of her show; our team produced a second long-form version and clips, then published them across Xiaohongshu, YouTube, Bilibili, and our community.
Also: 11.3K Xiaohongshu shares and 1,088 comments; 3.6K YouTube likes and 481 comments; 8,026 Bilibili saves.
“I hadn’t known her before, but she’s incredibly cool. I hope they talk every year so we can see how her thinking changes.” — YouTube viewer (translated)Watch the YouTube cut and comments
Counts come from public platform pages and screenshots captured in July 2026. Platforms use different counting methods.
HOW I WORK
ANOTHER SAMPLE · I AM THE HOST
This nearly three-hour conversation moves from intelligence, learning, and consciousness to embodied AI, brain–computer interfaces, and education. It has reached 130K+ YouTube views and 460+ comments.
Watch the full conversationYou choose the questions, pacing, title, and final edit. We can align on direction before recording; I do not require pre-approval of questions or the final cut.
I can bring primary sources, data, first-hand cases, and the strongest counterargument I can find—not scripted answers.
I will state the view clearly and say what evidence could overturn it. The best conversations make me re-check—and sometimes change—how I think.
Before release, I can help check names, dates, quotations, and data. The editorial conclusion remains yours.
For a jointly published episode, we can agree in advance on the primary cut, editing responsibility, clip rights, and release timing.
RECORDING DETAILS
Languages
Mandarin or English
Format
Remote, or in person in Seattle
Length
Usually 60–120 minutes; adaptable to the show
Based in
Seattle · Pacific Time
COMPLETE ARCHIVE
The full Xiaoyuzhou archive is preserved here for producers who want to check topics, formats, and earlier work.
屠龙大实话
59. 线性思维害死人？和康奈尔大学经济学博士课代表立正扒一扒这个思维中的隐形巨坑
十字路口 Crossing
高手怎么用 AI？普通人怎么学 AI？投资人如何投 AI？｜对谈课代表立正
The Build Log
#9 Is power at work an illusion? Borrowed authority and real influence
Grad Lounge 不止学术
118. From the ivory tower to real capability: building personal value in the AI era
Talk to JANE｜对话身边的人
拆解职场进阶的世纪谎言｜Waking Up at the End of the Ladder
AI炼金术
课代表立正：AI 课卖 2000 刀，偏偏坚持古法手搓
The Wanderers 流浪者
从焦虑到上手：白领 / 投资者的 AI 实战第一课【串台课代表立正】
INDIGO TALK
Quit ChatGPT to become an AI orchestrator / AI as a second Renaissance · EP46
人民公园说AI
腾讯“血脉觉醒”：马化腾朋友圈背后，小龙虾和大模型我都要！
创见｜对话顶级创投大脑
#031 关于 OpenClaw 被封杀以及我们为什么不建议普通人养龙虾？
牛油果烤面包
#148. 我们为什么 AI 焦虑，我们又该怎么脚踏实地
在路上｜ON THE ROAD
Vol.74｜大厂的收入是幻象，你的焦虑也是
破壁圆桌
EP14：对话课代表立正：AI时代下，后悔没早点离开职场
小火车商业进化论
课代表立正：知识内容大 V 分享 6 个你不知道的涨粉秘诀
StellaxAmy·自定义
EP77 Startup 选择、AI Evals 怎么做、Data Science 未来
佐治亚小帅｜北美华人故事访谈
S7E10 AI时代的新毕业生怎么办｜课代表立正
硅谷101
E195 · From tool to partner: seven experienced AI agent users
The Build Log
#1 课代表立正：打开第四面墙，走进课代表的世界
INDIGO TALK
对谈课代表立正：AI 时代生存指南 · EP20
王路在隐身
康奈尔博士、数据科学家、玄学、阿毗达磨
商业 WHY 酱
S3E03 · Before asking how to profit, ask the basic questions about ChatGPT
What’s Next｜科技早知道
How should we respond to ChatGPT? Highlights from a closed-door market discussion
随机漫谈 Random Talk
跨境电商都是怎么赚钱的？
HOST KIT
Use whichever bio length fits the show. The facts and links are here for verification, not as required talking points.
Download the six-page Chinese guest kit
1200 × 1200 headshot
Download headshotShort bio
Yuzheng Sun has a PhD in Economics from Cornell and is the founder of Superlinear Academy and AI Builders. He has worked at Amazon, Meta, Tencent, and Statsig, which was later acquired by OpenAI. His defining idea is MAKE WHAT LASTS: turn judgment and craft into work that remains useful after the first result.
Yuzheng Sun (课代表立正) has a PhD in Economics from Cornell and is a founder, operator, and author based in Seattle. He has worked as an economist at Amazon, a data scientist at Meta, Vice Director of Data & AI at Tencent IEG, and Principal Data Scientist and sole evangelist at Statsig, which was later acquired by OpenAI. He is the founder of Superlinear Academy and AI Builders. His defining idea is MAKE WHAT LASTS: turn judgment and craft into work that remains useful after the first result and unmistakably bears its maker’s choices. As of July 2026, he has taught 3,000+ paying learners and held 200+ public conversations with researchers, founders, and operators. He is co-author of Growth Data Analytics Playbook and author of 《真本事》.
One line for the introduction
Superlinear Academy stands for one idea: MAKE WHAT LASTS.
PROGRAM INVITATIONS