Technical writing on AI/ML, LLM systems, and production engineering.
Architecture walkthrough of a 4-node LangGraph pipeline for forensic interviewing — from fine-tuning Phi-4 to achieving 10/10 PEACE compliance at ~750ms latency.
A hands-on guide to parameter-efficient fine-tuning — from dataset preparation and LoRA configuration to training loops, evaluation, and production deployment.
How I built an end-to-end ad trafficking automation system using Vertex AI Agent Engine, Cloud Functions, and human-in-the-loop approval workflows.
Practical guidance on building retrieval-augmented generation systems that actually work — chunking strategies, vector DB selection, retrieval quality, and prompt engineering.
Reflections on 7+ years in AI engineering — from classical ML at enterprise scale to building LLM-powered agent systems.