Portfolio

Building a Paged KV Cache Allocator with Continuous Batching in Pure Python

This project demonstrates memory management, concurrent request scheduling, and low-level allocator design—all core skills for backend and ML infrastructure roles.

Build a Pure-Python GGUF and Safetensors Weight Loader

A hands-on guide to building a pure-Python gguf and safetensors weight loader from scratch. Parse binary headers, memory-map tensor data, and convert to numpy arrays for inference — with real runnable code and a roadmap to production-grade.

A schematic of a mixture-of-experts model routing tokens to experts.

Build a Minimal Mixture-of-Experts Inference Engine in Pure Python

Implement a minimal mixture-of-experts inference engine from scratch in Python, demonstrating systems thinking, custom loss functions, and production-ready patterns.

LoRA fine-tuning engine architecture diagram showing adapter layers, fused updates, and runtime swapping

Build a LoRA Fine-Tuning & Inference Engine in Pure PyTorch

A hands-on build guide for a LoRA fine-tuning and inference engine featuring low-rank adapter gradients, fused weight updates, and runtime adapter swapping in pure PyTorch.

A code editor showing a Python sampler class with visualization of probability distributions

Build an LLM Decoding Sampler from Scratch: A Portfolio Project That Signals Real Systems Skill

A hands-on guide to building a from-scratch LLM decoding sampler implementing temperature, top-k, top-p, and contrastive search in pure Python. Includes architecture, complete code, and a roadmap to senior-level extensions.