Crypto Intelligence that Doesn’t Suck: About a New Project I’ve Been Working On
I rarely use this blog to write about my work but I am going to make an exception today to write a small note about one of the most intellectually challenging and yet rewarding projects I’ve ever worked on.
These days, I divide my time working on both blockchain and artificial intelligence(AI) which I consider the most fascinating technology trends of the current generation. AI is redefining software as we know it by making knowledge an intrinsic component of the stack. Blockchain is powering the first financial asset class in several generation and one that is natively digital. A few months ago, we saw an opportunity to combine those two passions into a new project that leverages true data science to provide intelligence about crypto-assets. The result of that effort is the IntoTheBlock platform which we are finally unveiling in a Beta version today.
Crypto-asset are quickly becoming a relevant asset class in the financial ecosystem with market caps hovering on the hundreds of millions. Despite its size and growing ecosystem, crypto-assets lack a layer of market intelligence that provide clear insights about the behavior of this new asset class. Don’t take me wrong, the crypto market already has several analytic platforms but those are mostly constrained to some basic metrics such as counting transactions and address that only represent a fraction of the behavior of a crypto assets. Similarly, 10 years after the Bitcoin whitepaper we are still analyzing crypto assets using fundamental and technical analysis models that have very little correlation to crypto-assets.
As a new asset class, crypto-assets have new fundamentals and those require a new form of analysis. The most differentiated characteristic of crypto-assets is that they record the behavior of individual investors in public blockchains. Effectively, these blockchains are a crown jewel of information about the behavior of crypto-assets. From that perspective, if we can extract intelligence from blockchain datasets and complement it with off-chain data we have a very unique shot at better understanding and predicting the behavior of cryptocurrencies and tokens. Those two key principles: the fact that crypto-assets require a new for of analysis and that this new analysis should start with data science at the blockchain level were the genesis of IntoTheBlock. In the next few days, I am going to be writing a lot about the thesis behind this new platform but I wanted to highlight a few fundamental ideas:
What Did We Build?
IntoTheBlock is a market intelligence platform for crypto assets completely based on machine learning. The platform runs over a dozen of advanced statistical and machine learning model on an infrastructure that includes clones of the top blockchains in the market near real time and several off-chain data sources. Those machine learning models produce over 30 intelligent indicators and signals that are very unique to the crypto space.
We’ve assembled a phenomenal team of close to 15 data scientists, machine learning engineers, infrastructure/blockchain developers, business development that are obsessed with bringing a new level of intelligence to the crypto market space.
The initial feedback and market validation so far has been phenomenal but we need more. Please go and sign for IntoTheBlock BETA and send us your feedback.