Welcome to Datai Network

As a developer, you can contribute to the network by adding projections to increase data coverage and therefore earning $Datai tokens for your work. You can also source data from the network to build your own DApp or project.

Our extensive, step-by-step documentation equips you with the tools you need to get started. In the future, we will also open our grant programs that you can apply to. Stay Tuned!

Curious about how it works? Keep reading to find out!

If you’re here to source data instead, please jump to the Data Consumer section to explore more.

Note

In the current version, the only supported projection type is DeFi Active Positions. This is due to the fact that the Datai Network is in its early stages. More advanced use cases and projection types are planned in the roadmap and will arrive in the future versions. Read more in Available Projections.

What’s a Projection?

A projection is a bundle of code implemented by a developer who wants to integrate a specific protocol into the Datai Network. The Datai engine compiles these projections and deploys them to IPFS for indexers to run on their nodes.

What information does each Projection provide?

Each projection will provide the following information about a user position in a specific protocol:

• Supplied / borrowed token balances → Amount of tokens which were supplied to the protocol or borrowed by the protocol
• Position token balance → Synthetic token received from the protocol when creating a position (i.e. Liquidity Pool token)
• Reward token balances → Amount of tokens which are pending to be claimed by the user as rewards
• Pool Address → Contract address where the user is supplying or borrowing (i.e. staking contract or lending pool contract)

What are the components of a Projection?

A projection has two main components which will contribute to calculate the active positions previously explained:

• Subgraph → Indexes relevant events, such as a user staking tokens in a contract. The subgraph stores data about these positions, identifying users with open positions. This helps the next component focus only on these users rather than analyzing the entire user base.
• Watcher → Periodically analyzes positions stored by the subgraph and calculates the active position balances.