Machine-Learning

A visualization of a sparse Mixture-of-Experts routing architecture with multiple expert paths and a gating network.

Build a Sparse MoE Router from Scratch in NumPy

A hands-on guide to implementing a sparse MoE router from scratch in NumPy, covering noisy top-k gating, capacity-factor batching, and auxiliary load-balancing loss — a CV-worthy side project for ML systems engineers.

A neural network weight matrix visualized as a heatmap.

From Scratch: GPTQ‑Style Post‑Training Quantizer in NumPy

This guide walks you through implementing a GPTQ-style post-training quantizer entirely in NumPy. The resulting script compresses transformer weights and measures accuracy impact, showcasing your ability to ship low-level ML infrastructure.

From Scratch Sparse MoE Inference Engine

Build a sparse Mixture‑of‑Experts inference engine from scratch in Python, mastering top‑k gating, capacity‑constrained dispatch, and load‑balancing loss.

Conceptual diagram of a dynamic context buffer

Dynamic Context Buffer with Learned Relevance Predictor

This post walks you through implementing a token‑aware context buffer that learns relevance and prunes low‑value tokens to fit a target window.

Building a GGUF Weight Loader from Scratch in Pure Python

A hands-on guide to building a zero-dependency GGUF weight loader from scratch in Python, covering the binary container format, tensor extraction, and a roadmap to production-grade features.