Welcome to my corner of the web, a place for the things I build, the ideas I’m exploring, the lessons I’ve learned, the curiosities I keep chasing, and the art of making complex things work.
Ubuntu 24.04.3 LTS · x86_64
Last login: ... on pts/0
Type 'help' for available commands.
; profile.asm — identity loaded into memory
section .data
section .text
RET ; exit 0x00
process ready · interactive navigation attached
Architecture notes, leadership thinking, and things I learned the hard way.
A 30B model with 3B active parameters still needs room for its full weights. Here is how dense and mixture-of-experts models differ and what total parameters, routing, memory, and compute mean for local AI.
Tokens per second is not one measurement. Prefill is the work of reading the prompt; decode is the work of generating output. Understanding both explains TTFT, output speed, bandwidth, compute, context length, and benchmark results.
A base model, an instruct fine-tune, a LoRA adapter, a Safetensors checkpoint, and a GGUF file describe different layers of the AI model stack. Here is how they fit together and which one to download.
Things I build, break, and benchmark outside of work. Ongoing experimentation, not a polished portfolio.
A local coding benchmark for ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF at IQ3_S-MTP on an NVIDIA RTX 5080 with 16 GB of VRAM. Three single-file web builds test responsive styling, animated canvas generation, and interactive UI logic while recording the exact inference configuration and decode performance.
An interactive capacity-planning tool for local LLM inference. It estimates model weight storage, KV cache growth, runtime VRAM, minimum host memory, and GPU count from model size, quantization, context length, cache precision, batch size, and the target memory architecture. The breakdown makes the assumptions visible so a hardware plan can be adjusted before downloading a model or buying a GPU.
An automated cocktail-mixing machine built with nine 12V pumps, supporting up to ten liquors and mixers through a custom PCB, 2020 aluminum extrusion frame, and 3D-printed enclosure. A companion web app can discover and connect to the machine over Wi-Fi, manage ingredients, and prepare hundreds of cocktails with flexible recipe modifications and customization. The project brings together accurate dispensing, pump control, plumbing, wireless connectivity, and a software-driven recipe system in one programmable bar platform.
Three purpose-built systems, backed by a compact lab and local AI stack.