A place for developers to talk about building on Farcaster.

/dev

There’s a lot of noise and confusion around what AI can achieve in web2, and it’s even worse in web3. So, is there something really useful beyond useless llm agents with crypto-bro persona, built to shill useless coins ? 🤔

In my opinion, one of the best use cases for AI in web3 is mitigating traders biases and saving them precious time. There are already a lot of ways to automate trading and delegate time-consuming tasks, and today we’re going to explore one possible way using AI generated vector embeddings. 🤖

For the past few months I’ve been experimenting and trying to answer this question: Can we leverage vector embeddings to automate trading decisions ?

As a bit of context, vector embeddings are lists of numbers that represent words, sentences, images, or other data in a way computers can understand. Instead of seeing the word "dog" as just letters, the computer sees it as something like [0.2, -0.5, 0.7, 0.1...] - a list of numbers that captures its meaning.

This is how Spotify can suggests a podcast called "Feline mysteries and you" when you just typed "cat podcast". 🐈

How does this help us with market analysis ?

NodeJS | TS | Solidity builder 🔵
I like to experiment with FC frames 🔧

Imagine a trader documenting market observations daily for 5 years, converting each analysis into vector embeddings. When we analyze today's market and transform this analysis into a vector, we can instantly search this historical dataset to find the most similar past market conditions using cosine similarity search.

This search measures the angle between vector representations, not their size - values closer to 1 indicate higher similarity. This means similar market patterns are identified regardless of overall market scale.

By providing any new market observation, the system returns historically similar situations based on semantic meaning, not just surface-level metrics. This would allows the system to learn from past trading decisions made in genuinely comparable market conditions.

The key benefit is that it wouldn’t just match simple metrics like "prices dropped 5%" but deeper patterns like "altcoins falling while BTC dominance rises during regulatory uncertainty," enabling much more sophisticated pattern recognition that we would overlook or that we’re not aware of.

That said, actual embedding-ready text observations should avoid specific token names and all prices should be normalized to prevent bias in the similarity matching.

Normalizing removes the scale difference between assets (a $3,000 drop in BTC is very different from a $3,000 drop in a $5,000 token), makes historical comparisons possible regardless of price levels at different times and focuses the analysis on the pattern of movement rather than absolute numbers.

For the past months I’ve been experimenting with this concept and I have a system running on a daily basis on a VPS.

In the next post I’ll share the interesting results I’ve got so far, what I’ve done to improve it and some implementation details.
Also I'm very interested in hearing your thoughts on this.

Thanks for reading, have a good one 👍