// 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.
Build a GPTQ Quantizer with Layerwise Weight Reconstruction in Pure Python
Implement a GPTQ quantizer in pure Python, reconstruct weights layer by layer, and learn how this project signals expertise to employers.
Pure-Python Continuous Batching LLM Serving Engine with Dynamic Slot Allocation and KV‑Cache Management
Build a minimal yet functional LLM serving engine in Python that demonstrates continuous batching, dynamic slot allocation, and KV‑cache management—ideal for standing out to hiring managers.
Build a From‑Scratch Python Decoding Library with Temperature Scaling, Top‑k, Top‑p and Min‑p Sampling
A practical, from‑scratch decoding library that implements temperature scaling, top‑k, top‑p and min‑p sampling with a pluggable processor pipeline, perfect for demonstrating systems‑level Python skills in interviews or side‑projects.
Hands-On Build Guide: A GPT‑2 Style BPE Tokenizer From Scratch in Python
A practical, runnable Python implementation of a GPT‑2 compatible BPE tokenizer, complete with regex pre‑tokenization, merge‑rank encoding, and byte‑level fallback for lossless Unicode handling.
Building a LoRA Adapter with Fused Low-Rank Projections and Weight Merging in Pure Python
A practical guide to implementing a LoRA adapter with fused low-rank projections and weight merging in pure Python, highlighting systems skills for engineering roles.