Mar 13
Loading video player...
We are working on our blockchain project, building it from scratch in Rust and handling our own ledger. Today, the goal is to add a transaction to our database, which means setting up a simple write operation without worrying yet about adding signatures or complete transaction details. We use SQLite with the RustQLite wrapper for easy database access and define tables for transactions, with a plan to add a blocks table later. For quantum safety, we use Sphincs for our signatures, which are pretty large, but that is expected for this type of cryptography. Our smart contracts will run with Lua, and everything is designed for quantum resistance. The ledger setup could have been JSON, but we stuck with SQLite to allow for better indexing and querying. The client side will sync with the ledger server, but ultimately, the vision is a peer to peer system with a bootstrapping or seed server for the start, eventually becoming more decentralized. When it comes to code, we cleaned up function names, built simple query and insert operations, and debugged type issues with Rust’s results and SQLite mapping. There were a few hiccups with mismatched column names and data types, but after adjusting the schema and queries, along with creating a struct for transaction rows, we managed to write and read transactions to and from the database. On the AI side, we looked into adding machine learning abilities, checking out frameworks like Linfa, SmartCore, and especially Burn for deep learning in Rust, which looks promising for running AI tasks on blockchain data. After integrating Burn as a dependency, we planned future work for both AI tasks and blockchain improvements, including anti 51 percent measures.