I fine-tuned all-MiniLM-L6-v2 on dating profiles, flipped Spearman from -0.22 to +0.85, and found LoRA hit 96.2% of that with 0.32% of parameters.
Grid search across 16 RAG configurations reveals embedding model selection drives 26% more retrieval quality than chunk tuning.
My first LLM judge had a 0% failure rate. That meant it was useless. This is the story of calibrating it to actually catch failures, and building a correction loop that took synthetic data failures from 36 to zero.
Why I am building 9 AI systems from scratch while working full-time as an Engineering Manager. The portfolio, the progression, and what I have learned so far.