Monitoring social media text for potential cases of adverse events is an important, but manual and expensive task in pharmacovigilance. AE Brain uses an ensemble of NLP models (BERT, BioBERT, character-based models) to automatically detect potential mentions of adverse events and reduce the manual workload required for monitoring. By adopting a human-in-the-loop approach and rendering re-training fully reproducible our system ensurescontinuous improvement of our model and achieves GxP compliance.