$dreams framework: multi-chain ai agents sharing successful paths instead of learning from scratch
chain of thought processing + deepseeks-r1 reinforcement
DeSci FOMO is fascinating in this memecoin-driven market.
On one hand, DeSci is the exact opposite of memecoins:
"Boring science" vs. memetic cute dogs and cats...
And DeSci tokens attach more utility value than any DeFi token.
I thought utility was just an impediment and a ceiling for massive token pumps?
For example, $RIF or $URO grant fractional rights to the cocktail and could be licensed with potential revenue sharing.
But it will take years for this to materialize, even if the experiments are successful.
DeSci feels like the 2017 ICO era with ambitious ideas to change the world.
But this time we are actually doing it! It's not just a whitepaper.
Still, DeSci isn't a totally new narrative. We've heard about it years ago.
The speculative triggers for DeSci seem to be:
• Vitalik/CZ & Brian Armstrong are active advocates
• Binance recent investment in Bio Protocol & Brian being a cofounder at @ResearchHub
• Degens are actively looking for the "next big thing" in the bullish market
• Pumpdotscience launched tokens in pumpdotfun memecoin style
The narrative formation often seems random and takes a few triggers for them to emerge.
But few could have expected such FOMO for DeSci...
Still, DeSci is a redemption trade for all our degen sins.
Hope the FOMO lasts for a while now.
Memecoins aren’t about Gambling.
Memecoins are about Survival.
Young people have been stripped of all opportunity.
So they are inventing a New Asset Class.
Memecoins are here to stay and this is just the beginning.
Tip of the AI Iceberg: $ai16z
I’ve been playing around with mental models on how AI could gain mindshare the fastest.
The truth is, we’re currently in a glorified chatbot meta—an evolution of what we’ve seen with ChatGPT, just with some major upgrades. While agents like Truth Terminal, Zerebro and Cents are great, they haven’t drifted too far from the established path.
The next billion-dollar project won’t be just a better version of Truth Terminal. It’ll be something that steps outside the bounds of traditional chatbot KOLs entirely.
That’s when the concept of an AI trading agent hit me as the ultimate way to capture external attention.
I was always bullish on the idea since ai16z first launched but skeptical about the product given the complexity and development hurdles—until I dug deeper into their recent progress.
Turns out, they’re far ahead of the curve, with their first live trading session set for Friday (Nov 15).
I wouldn’t be surprised if a certain major fund is already watching closely. There has already been internal testing this week for smaller trades, but this upcoming livestream on Friday will be where things kick off.
ai16z sparks a different kind of imagination as it positions itself as the “Wall Street beater.” We’ve all dreamed of having a perfect algorithm making money while we sleep—this could be it.
Ask yourself: what captures people’s minds more? A Twitter chatbot with personality or an “infinite” money-printing AI bot?
Imagine a social analysis tool where an AI could monitor top traders in a Telegram group, continuously learning and refining its strategy based on the best calls. Eventually, this bot would aggregate the insights of these top traders.
Traders/KOLs could license or sell access to the bot based on their trading reputation, introducing a new level of “copy trading on steroids.” It would transform the KOL space by naturally filtering out those without genuine alpha.
For anyone paying attention in the teapot community (AI researchers), you know @shawmakesmagic is the most based dev out there.
He’s been letting everyone use his open-source software (Eliza) to run 90% of agents in AI memes, and he’s also a DeFi summer veteran, giving him the edge in crypto’s chaos.
Open-source development has always been a hard way of earning compared to private work, but Eliza’s framework and packages are changing devs perceptions. They enable indie devs to gain retroactive funding for public goods based on usage, crucial for onboarding the best AI talent.
His deep crypto and DeFi experience is ensuring that ai16z’s trading bot has built-in safeguards for common crypto landmines.
This trusted execution environment (TEE) acts as a social analysis tool, monitoring top traders and adapting strategies accordingly.
Another product vision is something like Pump Fun for AI trading agents: a marketplace where people create, backtest, and launch custom trading formulas that others can invest in.
This marketplace will allow devs to sell bot actions (scripts), providing better access to tools for quant processes. Think of it as a storefront for AI trading enhancements, giving developers the chance to make their unique strategies widely accessible.
A big advantage of bots in today’s market is their unemotional approach to trading—ideal for navigating the information overload. Bots can’t solve every problem in crypto, but they can provide a better hit rate with constant automatic scanning.
As the DAO builds what could be the Pump Fun of AI trading agents, they’re aware of the risks of easy deployments. These bots need careful tuning, and allowing freedom of deployment too soon could lead to fraud, similar to what we’ve seen with Pump Fun tokens.
They’re also working an agent accelerator where agents built on the DAO’s framework can send a percentage of tokens to the DAO in exchange for more exposure if legitimate and can work in tandem with the ai16z ecosystem.
Whoever claims the first-mover advantage in the AI Agent marketplace will see a dominance comparable to Pump Fun’s in memecoins.
If you can get an entry in the 40 mil range I think this is a good r:r going into Friday’s stream, if all goes well I’m expecting the following week to be filled with a lot of new noise about the progress and potential success of this narrative.