Technical breakdowns
Building a SaaS platform that imports your Chess.com games, analyzes every position with Stockfish 16, and creates personalized puzzles from your blunders.
A deep dive into creating a real-time AI system that detects units, classifies types, and estimates tower health from gameplay footage.
How I built a platform that engaged 500+ community contributors to create a massive dataset for training computer vision models.
The story behind building a chess tactics trainer that uses spaced repetition and the Lichess API to help players improve.
How I developed a proof-of-concept for offloading ML computation to the cloud for real-time gesture recognition in automotive applications.
Building an automation bot that handles complete Clash Royale gameplay using PyTorch and OpenCV for competitive play.
The journey of building an AI-powered World of Warcraft fishing bot with YOLO, and launching it as a commercial product.
Industry perspectives
A product-driven argument for restraint, clarity, and saying no when your user base starts expanding. Growth demands discipline, not expansion.
How ChessPecker grew from a personal tool into a 1,300-user SaaS and why freemium was the most controversial but necessary decision I've made.
Learn how to build scalable real-time data pipelines for AI systems using PostgreSQL, Supabase, and WebSockets. A practical guide to streaming analytics and ETL automation.
A hands-on guide to building autonomous AI agents with LangChain and modern frameworks. Learn agent architecture, implementation patterns, and reliability strategies.
Edge AI leads the next wave of intelligent systems. Explore how on-device inference transforms latency, privacy, and scalability for modern ML applications.
After using GPT-4, Claude, and Copilot daily for months, here's what actually works and what doesn't in real production environments.
Everyone talks about edge computing for ML, but the reality is more nuanced. Here's what 3 years of building vision systems taught me.
The framework wars are over, but not in the way most people expected. A practical take from someone who uses both daily.