ML Data is live. This site publishes price data for crypto assets, read from the decentralized exchange pools where those assets actually trade. This first post is the introduction: who we are, what stands here today, why we built it this way.
What ML Data is
Every number on this site is computed the same way. Find the pools where a pair changes hands, across every chain we observe. Weigh each pool by the liquidity it carries. Filter the readings that do not belong. Pin the result to a block, so the price is a fact about a moment rather than a rumor about a market.
The result is a market rate with its origin attached. Most price services hand you a number; we hand you the number together with the market behind it: which pools, on which chains, at which weight, as of which block. If a reading surprises you, you can open the composition behind it and check it against the chain yourself. Nothing on a feed page asks to be trusted blind.
Who is behind it
ML Data is operated by Matrixedlabs Technologies - FZCO, whose information security management system is certified to ISO/IEC 27001:2022 (certificate details).
The underlying blockchain infrastructure is operated by Matrixedlink B.V. (Matrixed.Link), an affiliated company that has operated validators plus oracle infrastructure since 2021. Its team runs more than 500 live price feeds in production; StakingRewards rates it AAA as an operator. ML Data’s first client is the Chainlink ecosystem; the same feeds are now available to anyone whose work depends on accurate prices.
We are operators before anything else. That shows in the product in small ways: data statuses that say what they mean instead of dressing a quiet market up as a fresh one, a catalog that includes thin markets because completeness is part of the job, pages that explain their own readings.
What is live today
The inventory, with the proof one click away:
The catalog. More than 500 price pairs across more than 130 base assets at launch, grouped into eight categories: stablecoins, liquid staking tokens, wrapped bitcoin, yield-bearing assets, euro stablecoins, Solana staking tokens, DeFi tokens plus the majors they all trade against. The full index lives on the coverage page, searchable by symbol.
Feed pages. Each shipped pair has a page of its own showing the live rate, the pool composition behind it, depth within one percent of the mid, plus the honest status of the reading. Start with BTC/USD or stETH/USD to see the format.
Live boards. Five boards read the same data sideways: wrapped bitcoin against BTC on the BTC parity board, staked ether against ETH on the ETH ratio board, Solana staking tokens against SOL on the SOL ratio board, stablecoins against each other on the dollar cross board, computed pair rates on the derived cross board.
The API. Everything above is a view on one data layer; that layer is available over REST for teams that want the numbers inside their own systems.
Why we read the pools
DeFi settles on-chain. Liquidations clear against pools. Collateral is worth what it can actually be sold for, where it can actually be sold. For a growing class of assets, stablecoins born in DeFi, staking tokens, tokenized funds, the on-chain pools are not a secondary market but the market. A price layer for that world should stand where settlement happens. It should also be checkable by the people who rely on it.
The longer version of that argument, told from the engine room, is in Why we built a DEX-native price feed.
Who it is for
Three kinds of teams told us what to build. Protocols that liquidate against prices and answer to their governance for the data they choose. Trading desks that mark positions and manage risk on assets whose real market lives on-chain. Institutions that need a verifiable answer to the question “where did this number come from” before a feed enters their stack. The methodology page documents the method end to end; the enterprise page covers procurement.
What comes next
The catalog keeps growing: pairs enter as their markets earn a place in the composition; shipped pages are reviewed one by one rather than generated in bulk. This blog will follow a steady rhythm of one article at a time, starting with the story behind the product and a guide to reading our numbers.
If your work depends on prices for the assets we cover, talk to us. We would rather show the data than describe it.
