Building an LLM System from Scratch in Pure Python & NumPy: Architecture, Invariants, and Clean Code

Hi everyone,

I wanted to share the design invariants and architecture of Draco AI, a full-stack, hardware-agnostic LLM system built completely from scratch in pure Python and NumPy. The core philosophy behind this project is strict code quality, mathematical correctness, and clean architecture—completely eliminating heavy framework dependencies like PyTorch or HuggingFace.

The baseline includes a robust pytest validation suite covering GQA, MLA, hybrid attention, Medusa heads, and speculative tree decoding execution paths.

GitHub Repository: https://github.com/DUCNGUYEN-creator/Draco_AI

Discord Community: https://discord.gg/dJXMbexdTh

Building an LLM System from Scratch in Pure Python & NumPy: Architecture, Invariants, and Clean Code

Related reading

I built an interactive 11-chapter guide to how LLM inference actually works

LLM APIs as Infrastructure: Building Deterministic Systems Around Probabilistic…

How I Fixed LLM Hallucinations on a 512MB Server with Pure Math

I built an LLM eval framework from scratch. Here is what I wish I had bought…

I Built an LLM Gateway That Extends Claude Pro/Max Users with Azure AI Foundry,…

Running LLMs Locally in 2026: The Complete Guide to Benefits, Trade-offs, and…