Blog

Findings from building AI systems. Evidence-based, not opinion-driven.

Fine-tuning Embeddings Evaluation

The Model Wasn't Random — It Was Backwards

I expected a pre-trained embedding model to be neutral on dating compatibility. Instead, it was systematically wrong — scoring incompatible pairs higher than compatible ones. Here's why that made fine-tuning more interesting.

5 min read

RAG Reranking Evaluation

Reranking Isn't Always Better: A Tale of Two RAG Systems

I built two RAG pipelines. In one, adding a reranker dropped accuracy by 7%. In the other, it improved accuracy by 7%. Same technique, opposite results.

6 min read