// TODO: I’m martinuke0

Welcome to my corner of the internet. This website is a personal blog which I use as a platform to document my learning journey and showcase it for the world to see.

Hands-On Build Guide: Minimal Mixture-of-Experts Routing Layer for Transformers in Pure Python

A step‑by‑step guide to implementing a minimal MoE routing layer from scratch, using only NumPy and pure Python, perfect for a portfolio CV.

Build a Pure‑Python LLM Sampler: Temperature, Top‑k, and Top‑p from Scratch

Learn to implement an LLM sampler from scratch in pure Python, controlling temperature, top‑k and top‑p, and add a standout project to your CV.

A schematic of attention mechanism with floating point arrows

Build a FlashAttention Kernel from Scratch with Python and NumPy

Implement a minimal yet functional FlashAttention kernel from scratch, leveraging NumPy’s array ops, and learn how to extend it for production use.

Diagram of a paged attention engine with KV cache blocks

From‑Scratch Paged Attention Engine with Variable‑Size KV Cache Blocks and Ghost Eviction

A hands‑on guide to implementing a paged attention engine in Python, including block management, ghost eviction, and testing.

Abstract illustration of network packets flowing through a kernel

Mastering eBPF: Implementing Kernel Probes for High-Throughput Packet Filtering

A practical guide to using eBPF kernel probes for high‑throughput packet filtering, with code examples and performance tips.

Feedback