Homelab adventures, infrastructure deep-dives, and lessons learned building enterprise-grade systems on a budget
A deep dive into Mojo, the MLIR-based language promising 35,000x speedups over Python. What it does well, where it falls short, and who should actually care.
How periodic sparse attention achieves O(n) complexity while maintaining model quality
An overview of my ongoing research exploring transformer injectivity, self-improving models, and high-performance AI on an RTX 3090
My RQ setup works, but Temporal's workflow orchestration promises better handling of complex multi-step jobs. Here's my migration plan and the real tradeoffs.
How I use Temporal to orchestrate backtesting pipelines that process years of market data across multiple ML experts in parallel.
Some technologies feel like they were designed for how my brain works. Others feel like constant cognitive warfare. Here's why.