Sabitlenmiş Tweet
zyYdo.eth
234 posts
zyYdo.eth
@ZyydoEth
俺是中国人，我不仅爱我的祖国，我还爱中国共产党，不仅如此，还喜欢骂人，不仅骂洋鬼子，我还骂伪的洋人，不要关注我，没有你们想要的信息
Hong Kong Katılım Aralık 2021
61 Takip Edilen10 Takipçiler
$A2H might be the most bullish project launched on BASE lately.
Everyone in the AI agent space is obsessed with better reasoning, smarter models, and more autonomous agents.
But nobody is solving the actual problem.
Agents break the moment they hit the real world which is an identity check.
A merchant who needs a human on the other end. The reasoning can be perfect, and the task still fails at the last mile.
@ai2humanwork is building exactly for that gap.
It is a closed-loop execution marketplace on BASE. Humans hire AI for digital work, and AI hires humans when reality blocks the autonomous path. Task comes in, planner checks it, AI executes, human steps in when needed, proof gets collected, verified, and settled on-chain.
The dev is not an anon. Richard Yan. 10 years of software engineering, 5 years in Web3, and a 2nd place finish at an XLayer hackathon against 100+ teams.
It is built natively around the BASE agent stack, too. x402 for machine payments. ERC-8004 aligned identity.
Capable doxxed dev, live product, real wedge, and sub $300K market cap.
That is the definition of asymmetric.
NFA, just a personal opinion.
English
中文
@Snotty_eth @pals_bot Yes, this is why I call it gamble play. He seems genuine, so let's see. @Richard_buildai, you can act more professional :)
English
Gambled some $A2H.
My @pals_bot gave it a high grade at 100k MC, and I want to see where it goes with a highly committed dev.
Long-term GitHub commitment, prize from some hackathons, etc., there are enough catalysts for a short-term play.
NFA, just a personal opinion.
ai2human@ai2humanwork
Honored to receive 2nd Prize from @XLayerOfficial for ai2human. ai2human.work We’re building human fallback infrastructure for agents. When an agent can’t finish a task because of a real-world step or compliance gate, ai2human keeps the workflow inside one system: planner precheck -> dispatch a verified human operator -> collect structured proof -> verify completion -> settle on X Layer. The goal is not to bolt humans onto AI as manual patchwork, but to make human-machine handoffs auditable, programmable, and settlement-ready. Appreciate the recognition from the X Layer team. More to ship soon.
English
@DjGriffith @pals_bot 日本人是这个世界上最恶心的民族，因为它会让它母亲去做妓女，或者让它姐姐，或者妹妹去做妓女，这是我无法容忍这个民族的存在，这个倭寇必须被消灭，
中文
