You've got the right idea--there's lots of recent literature treating the typical IRT problem as a Bayesian inference problem. Some of the references listed at

You've got the right idea--there's lots of recent literature treating the typical IRT problem as a Bayesian inference problem. Some of the references listed at the bottom of the article (especially the Kim one) provide great literature surveys.