Threads on the growth behind Chinese AI, robots & EVs 🦾 Web3 Marketing 转行做 AI 增长 · 把中国科技讲给世界

China
Joined February 2017
If agents do the work, do we still need task boards? I use Linear to track projects and assign tasks to different AI agents. Will tools like Codex and Claude build project management into their own interfaces? How are you keeping track of work across agents?
🤖 Made with AI
1
75
1/6 China may already be running the world’s largest mass-market AI entertainment experiment. 🤔 see the Hongguo threads below 👇
1
184
5/6 If you watch closely, you can see the ByteDance flywheel: Fanqie’s web-novel IP feeds AI and live-action production, including Seedance-powered content. Hongguo distributes it, Douyin drives discovery, and the resulting ad revenue funds more production.
1
35
6/6 China’s new rules took effect on Sept. 1. They require AI-generated content to be labeled and introduce algorithm audits aimed at reducing addiction and excessive spending.
13
If you watch closely, you will find the ByteDance loop: Hongguo ties together Fanqie web-novel IP, AI and live-action production powered by seedance, Douyin distribution, ads income for Hongguo. All done in house.
1
51
This new Chinese rules about labling AI content is active since Sept 1, requiring AI labels on every generated episode and algorithm audits aimed at addiction and overspending.
15
This framing matches my experience across brand, content, community and growth. AI-era marketers need four capabilities: marketing fundamentals, data and growth, systems building, and judgment and taste. Agents can run more of the workflow. We decide what deserves to scale.
The "marketing engineer" is the NEW forward deployed engineer, and I think the BEST ones will make $1M a year! A forward deployed engineer embeds with your team and uses AI to build the workflows The marketing engineer does that BUT for growth, they build AI agents that find your customers, write your outbound, test your ads, and get smarter every week. The most valuable marketer changes with EVERY major tech wave: 1. Traditional marketer: make people care with story (print, radio etc) 2. Digital marketer: the marketer who owned new digital channels (SEO, PPC) 3. Growth hacker: the growth marketer who lived in loops, PLG and retention 4. Marketing engineer: the marketer who builds AI agents that run the whole system. The type of agents a marketing engineer would create: 1. The customer language agent. It pulls your Gong call transcripts, your Intercom tickets, and your G2 reviews every week, extracts the exact words customers use to describe the problem, and drops a memo ranked by how often each phrase shows up. 2. The buying trigger agent. It watches for signals that someone's ready, a company posting a job for the role you sell to, a funding announcement, a competitor getting torched in a review, then enriches the contact through Apollo and drafts the outbound tied to that exact trigger the second it fires. 3. The SEO gap agent. It pulls keyword gaps from Ahrefs, checks who's already ranking on page one, reads the top three results, then writes a better post with the founder's actual take baked in, plus the meta title and internal links, and drops it in for approval. 4. The creative testing agent. It takes one offer, generates 100 ad variations across three different angles with Nano Banana and your copy model, pushes them live through the Meta API, and kills anything under a 1% CTR on its own so only the winners keep spending. AND MANY MORE. If you're a marketer, this is how you stop being replaceable. If you're a founder, this is how you grow your company efficiently. AI agents are here and marketers are about to have a field day!! I explain everything (tools, agents, 30d plan) on today's episode of @startupideaspod Watch: --> youtube.com/watch?v=8ZC1G1ez… Marketing engineers are here. Whatever we end up calling them, the top 1% of them are going to make an absurd amount of money.
29
The robot race just opened a new front: car factories. Xpeng's humanoid unit closed a $900M+ round at a $6.3B valuation — Tencent and Alibaba came in as strategic investors. Mass production: end of 2026.
1
40
The crowd is forming: Chery's robot-arm unit at a ¥2.5B post-angel valuation, GAC with ~50 robots on real projects, NIO's AD chief spun out into a physical-AI company.
1
9
Growth lens: these aren't demo studios, they're manufacturers with lines, margins and distribution. If the humanoid "ChatGPT moment" is 2-3 years out, the advantage goes to whoever can ramp at cost — and carmakers already buy actuators at auto volumes.
10
China's most valuable listed company isn't an AI lab — it's a memory chipmaker. CXMT (ChangXin Memory) just posted its first half-year as a public company: revenue ¥150.3B, +873.6% YoY; net profit ¥77.6B. A decade of accumulated losses, erased in six months.
1
47
DDR4 hit end-of-life just as every AI server started pulling DDR5/LPDDR5. Global DRAM went scarce, prices spiked — CXMT's gross margin hit 84.8%. Its share: 7.67% globally — #4 worldwide, #1 in China, double where it sat a year ago. Samsung + SK Hynix + Micron still hold 90%+.
1
90
The lesson: don't beat Samsung — be the credible marginal supplier when demand outruns the oligopoly. Its customers: Alibaba Cloud, ByteDance, Tencent — the AI capex wave itself. Caveat: DRAM is cyclical and the big three are expanding. A shortage's face, not a moat.
44
DeepSeek gave its model eyes. Vision was the last closed-model toll booth — the toll just went free. MIT weights · 305B MoE · beats Opus 4.8 on 3 agent benchmarks
32
Tencent just open-sourced a frontier-class model. 770B parameters. 1M-token context. Apache 2.0. Input price: $0.83 per 1M tokens. Hacker News talked about it all weekend.
1
44
The fun detail: Hy4 claims it helped build itself — a +31.8% throughput optimization in its own training pipeline. Self-improvement claims are unfalsifiable. Still a great headline.
1
24
And it wasn't alone this week: Qwen previewed its next architecture, GLM's 5.3-Flash went open. Three frontier-class open models from China in 72 hours. Cadence × price × open weights. That's the moat.
33