First Release of FELT Labs — Federated Learning Platform — Finally, it’s here! Today I will present the first version of the FELT Labs tool for federated learning on Ocean protocol. This one definitely took longer than we expected. However, this article isn’t about all bugs we had to overcome. …

Tutorial on how to use native MetaMask encryption-decryption functions — Every Ethereum account is associated with private/public key pair. A natural question is how to use these keys for encrypting data? Does MetaMask support encryption? This tutorial will show a user-friendly way to use MetaMask to exchange encrypted messages on a blockchain. Throughout this tutorial, I am using TypeScript notation…

Making federated learning even easier than before — We are thrilled to announce the latest release of our application, which includes several exciting features and enhancements. Over the past few months, we have been refining the app and adding new functionality to make it more powerful and user-friendly. This includes a Python interface, custom algorithms, user accounts, the…

Short update about what’s new and what’s coming next — New month, new release! After releasing the MVP of our new architecture, we are excited to add new features. With a working version underneath, rolling new versions is now easier, and more updates are coming in the near future. …

What we learned from building FELT Labs from the ground up — We went through many changes, not only renaming from FELToken to FELT Labs but also changing our architecture, target audience, and more. It’s crazy to think that it will be a whole year of work on FELT in a few months. …

Moving to the next stage of FELToken development It’s been a few months since we started with the FELToken project — a web3 tool for training machine learning models on decentralized data. Even though it might seem like we are way past starting line, recently, we faced a decision which…

Why we split our monorepo and how we handle smart contracts — First things first, I think it’s totally fine to put all code into one place, especially at the start of the project. Just get things done. Don’t overcomplicate it. However, eventually, things get complicated. Dealing with smart contracts is tricky because once deployed, they are out there, and updating isn’t…

Introduction to FELToken and its relation to Ocean protocol — Web3 projects like to use fancy terms to describe their functionality. Intentional or not, these terms often hide the true functionality of the project. We often do it as well. I like to blame it on tweet length limitations, using terms such as decentralized federated learning, distributed AI, privacy-preserving machine…

Train your first machine learning model on private data — The first big popularization of federated learning was made by Google when they used this approach for improving Google Keyboard suggestions. The key problem is that data is distributed across multiple devices which they can’t leave (e.g. for privacy reasons). However, we still want to train machine learning models on…