AI Engineer building AI systems, developer tools and open-source software.
Featured Work
all projects →
TextStruct
A local-first extraction pipeline that turns scanned textbooks into structured, LLM-ready data — layout-aware OCR, deterministic math cleanup, pedagogical role classification, and LanceDB output.

Tracium
A lightweight, cross-platform desktop network health monitor for home connections — continuous latency/jitter/loss tracking, speed tests, Wi-Fi diagnostics, bufferbloat grading, and exportable reports.
Work Experience
- October 2025–Present
AILACDS
AI Intern → Junior AI DeveloperBuilt RAG and document systems for education and tax workflows, with retrieval across a ~90 GB legal/tax corpus.
- January–July 2025
MarketsandMarkets
Technology InternBuilt a lead-classification system that reached approximately 89% accuracy on operational data.
From the Cookbook
all articles →Writing
- 2026-08-25Three Levels of RAG
- 2026-08-12Why Agentic Coding Needs CI/CD More Than Human Coding Does
- 2026-08-11Advanced RAG in Practice: Why Hierarchical Retrieval Beats Flat Search
Open Source
Current Focus
Outside of work I'm spending most of my time on the LLM Cookbook and SLM From Scratch. The Cookbook is where I organize everything I'm learning about modern language models, while SLM From Scratch is where I put those ideas into practice by training models on consumer hardware. I'm also building Tracium, a cross-platform network monitoring tool written in Rust.