Thanks everyone this is all very helpful, great community here. As an initial attempt to fix it I tried constraining gamma to be positive and estimating more priors but this didn’t seem to change much.
I think I found a potential solution here: Non convergence issue on polytomous IRT model
That said, I’m getting in to territory that’s over my head and need to turn my attention to other things at the moment. So I’ll take a few days to read up, test a few solutions, and post back here with what I find/follow up questions.