Daniel 다니엘

So i've been thinking about an LVR-aware analyst AI agent to trade in some base AMM pool. There is a fairly new paper i would like to try and implement. The most common pattern for on-chain trading agents is LLM + a price feed + a swap router. It's basically a system that rely on LLM output based on basic price data thrown in its context window, this is bad, very bad, every person with a grain of intelligence will tell you that. It's a cheap approach but will rarely give any kind of good result, naming your agent quant won't magically make it quant level. I think there is still a decently big gap for more rigorous and proven analysis patterns that an agent can use to trade. Would be a fun experiment, maybe won't turn out to be effective but would still be fun to implement as my first agent on base.