Fine-Tuning Was the Wrong Answer Three Times
14 May 2026, 10:15 – 10:45 · Room 2B · AI Engineering · Talk (30 min)
Three projects where fine-tuning was the obvious move and the wrong one, with the numbers that showed it. Retrieval beat it, then prompt structure beat it, then a smaller model with a better evaluation set beat it. The fourth project fine-tuned, and this is how we knew that one was different.
Speakers
Kenji Watanabe
ML Infrastructure Engineer, Orbital