Writing on AI, cryptography, distributed systems, and whatever else has my attention.
Articles, series, and a daily journal by truthxify — on what I'm building, exploring, teaching, or just thinking about out loud.
Featured Series
View all seriesLatest Articles
ArchiveMay 25, 2026 · 15 min
Implementing Decision Trees from Scratch
Build a decision tree classifier from scratch in pure NumPy — entropy and Gini impurity, information gain, recursive splitting, and evaluating the tree on a real binary classification dataset.
From series · ML From Zero
- May 25, 2026
Implementing a Multi-Layer Neural Network from Scratch
Build a multi-layer neural network from scratch in pure NumPy — activation functions (sigmoid, tanh, ReLU), forward propagation, backpropagation derived from first principles, and visualizing decision boundaries on linear and non-linear classification problems.
From series · ML From Zero
- Apr 30, 2026
Learn CKB in 45 Minutes: A Hands-On Guide from Zero to Deploying Your First Script
A single resource that takes you from zero CKB knowledge to writing, testing, and deploying a real lock script in Rust. 13 chapters, 40 questions, one working project.
From series · Building on CKB
- Oct 22, 2025
Polkadot Hub: The Smart Contract Center of Polkadot 2.0
Explore how Polkadot Hub unifies rEVM and PVM into a single smart contract platform, marking the evolution of Polkadot 2.0 toward a multichain execution layer.
From series · Polkadot From Scratch
- Oct 9, 2025
Building a Crowdfunding dApp with ink! Smart Contracts and PAPI/ReactiveDOT
The tutorial walks you through building InkFundMe, a decentralized crowdfunding dApp using ink! smart contracts (for a PSP22 token and campaign logic) and a React frontend powered by PAPI (Polkadot API) and ReactiveDOT. Users can create campaigns, contribute with tokens, and claim refunds if goals aren’t met. It covers contract architecture, security best practices, deployment to the Passet Hub testnet, frontend integration, and optional Vercel deployment, resulting in a fully functional, secure, and customizable crowdfunding platform on Polkadot.
From series · Polkadot From Scratch
From the Journal
Open journal- Jun 8, 2026Reinforcement Learning(RL) trains an agent to take actions in an environment, the agent does not get told what to action correct, it basically gets a reward signal instead.
- Jun 3, 2026Principal component analysis(PCA) allows us to take data with a lot of features say hundreds or thousands of features and reduce it to very small number of features say two or five features.
- Jun 2, 2026Recommender system predicts how a user would rank or respond to items they haven't seen yet, then surface with the highest predicted score
- Jun 1, 2026Anomaly detection looks at unlabeled dataset of normal examples, learns what normal looks like, then flags new examples that looks different
- May 26, 2026Clustering looks at a dataset and tries to find points that are similar to each other and group them together