Short essays, experiment logs, and distilled notes. Right now these are placeholders — as I formalize posts, I will link them here.
A practical look at where classic sparse methods beat dense embeddings, and how I design experiments to choose the right representation for a task.
Read (coming soon) →Notes from building agentic systems where retrievers, generators, and critics collaborate — and what actually matters in evaluation.
Read (coming soon) →How my work evolved from traditional machine learning and uplift modeling to building LLM-centric, agentic products.
Read (coming soon) →