Soda Protocol

Dashboard
Borrowing
Decentralized governance - Soda Club
Sol ID Demo

Inspiration

Asset utilization has large difference between protocols, which demonstrates a low on-chain capital efficiency. Also, a credit system to connect both users and those protocols has not been established.

What it does

Soda provide a easy-to-use lending protocol for DeFi users. With credit rating system Sol ID, Soda can help users build their credit model and profile by analysing their on-chain behaviours. Sol ID will incentivise users to accumulate good on-chain behaviours which can bring them a long-term right, such as a higher LTV and potential airdrop.
Soda can also served as a liquidity pool and leverage provider which offers under-collateralised loan to certified protocols and institutions to increase the overall capital efficiency in Defi ecosystem. Protocols can also acquire credit data in Sol ID to help them provide better service for their users and achieve a refinement operation.

How we built it

Taking Compound and AAVE as reference and combined with our knowledge and experience on the big data and machine learning, we create the credit system, Sol ID, on top of our basic lending functions.

Challenges we ran into

Stable access to on-chain data
Data analysis and credit model design
Realisation of under-collateralised loan
A good risk management and governance model design

Accomplishments that we're proud of

We have find a method to maximise the capital efficiency for DeFi ecosystem and a strategy to support our short-term and long-term business. Technically, we have achieved: 1. flash loan, 2. flash liquidation, 3. extendable slot limitation for lending & borrowing, 4. on-chain credit rating system (Sol ID).

What we learned

TVL is no longer the only index that matters, instead, capital efficiency becomes the key of DeFi 2.0.
Solana's unique account and on-chain program model requires us to focus on issues such as authentication for on-chain program.
Using Solana's high-concurrency engine, many applications with higher real-time requirements can be implemented on-chain.
For complex business logic, we need to split the instruction and transaction, which may destroy the atomicity. Under this situation, strict and careful analysis of whether there are vulnerabilities/bugs is necessary.
Optimization of the code to reduce the time and space complexity of the algorithm, to ensure the consumption of the compute unit is allowed.