Projects
Projects documented with architecture decisions, stack rationale, tradeoffs, and what I'd do differently.
Enterprise RAG · HR & Policy Management
A working RAG prototype that lets employees ask plain-English questions about company policies and get cited answers from source documents. Built specifically to show the data preparation problem most enterprise AI projects fail at — and how to solve it.
Stack
Voice AI · Service Business Automation
An AI voice receptionist that answers every call, books consultations, and captures leads 24/7 for service businesses. Handles full conversation flow, qualification, and calendar booking without human involvement.
Stack
Personal Finance · Consumer Product
A personal finance platform built to help everyday people get out of debt through education, calculators, and community. Built end-to-end as a real deployed product.
Stack
Local AI Infrastructure · Privacy-First RAG
A fully local RAG pipeline running on an M4 Mac Mini. No external API, no data leaving the machine. Built to understand what running AI on real hardware actually requires: the memory constraints, quantization tradeoffs, and architectural decisions that cloud-based development never forces you to confront.
Stack