@mesutgulecen

Some details about ACM (Agentic Capital Markets) Agents and what we’re cooking at @samterminalcom: + The problem: People prefer simple and automated solutions instead of setting up their own workflows. + The solution: We've built an AI model, trained with historical ecosystem token data, to manage funds for users. We started building the first ACM AI model using data from over 20 successful $clanker ecosystem tokens. We're training the model on 72 different onchain data points, including all price, holder, and smart wallet-related data. The goal is to ensure the AI model identifies common patterns that typically occur before a significant price move. The model finds these success patterns, classifies them, and begins scanning for new @clanker token launches. When new launches show similar patterns, ACM agents buy the token with an exit strategy. The results from these new positions feed back into the model, creating a feedback loop to improve and generate new success patterns. Some of the tokens we're using with their historical trade data include: $LUM, $i, $PIZZA, $CLANKER, $NOICE, $FAIR, $BNKR, $CODY, $EMERGE, $FARTCOIN, $SKY, $FANS, $RETAKE, $QR, $MACHINES, $BETR, $DICKBUTT, $ANON, $NATIVE, $checkr, $DRB, $WARP, $DIME, and more. The first Clanker ACM agents are currently live on @samterminalcom in prototype form. As we improve the model, we’ll continue updating the agents. In the coming weeks, we plan to launch new ACM agents for different Base markets or categories.

0 reply

0 recast

1 reaction