Ok, it looks like you are trying to do too much in one go, this is a multistep process so do one bit at a time. Nucleophilic attack of the N on the carbonyl is a good idea, that will come into play later, but is probably not the first thing that happens. First off, what do you know about the reactivity of alpha-bromo carbonyls (compared to basic alkyl bromides)? Also, what is the most nucleophilic centre in thiourea?