AI & ML

Model drift discussions happening but no tracking yet

First post and replies  | Last post by Aaron Siegel, 6 years ago

Multimodal sounded exciting but integration turned complex

First post and replies  | Last post by Stephanie Hyde, 6 years ago

Prompt versioning ignored earlier now causing confusion

First post and replies  | Last post by John Copening, 6 years ago

LLM evaluation still feels manual not scalable

First post and replies  | Last post by Melanie Claypool, 6 years ago

Accuracy drops sharply on domain specific queries

First post and replies  | Last post by Emily Mccormick, 6 years ago

Grounding via RAG helped but not completely eliminating hallucinations

First post and replies  | Last post by Dennis Tomlin, 6 years ago

Chatbot added another layer instead of reducing complexity

First post and replies  | Last post by Gary Chan, 6 years ago

ML pipelines look good on diagram messy in reality

First post and replies  | Last post by Chris Yi, 6 years ago

Not sure how often models should be retrained feels like guesswork

First post and replies  | Last post by Melissa Lasoff, 6 years ago

AI infra cost slowly creeping up month by month not sure where to cut first

First post and replies  | Last post by Tabare Caorsi, 6 years ago