Adoption of Machine Learning (ML) outside the field of research has been one of the key factors to fuel disruption in several industry verticals. ML is a powerful tool for learning to solve complex problems and is generating interest in the health care life science space (HCLS), which poses compelling questions to this technique. In particular, Deep Learning (DL) is amassing more and more interest due to the high performance it has delivered in other fields . However, Deep Learning is yet to become a comprehensive tool for all HCLS applications. This is primarily because of the peculiar nature of Healthcare. Unique challenges that clinical data bring to ML include but are not limited to multimodal and sparse nature, lack of pre-trained models, and compatibility with conventional statistics in a field that heavily relies on p-values. At Amazon we are well aware of the difficulties in designing ML products for medical applications . For instance, NLP services as Comprehend and Transcribe have branched off their main products to create Comprehend Medical and Transcribe Medical to overcome the limitations of their generic counterparts and abide to laws like HIPAA and COPPA . The Alexa team has solutions dedicated specifically for HealthCare.. Examples of this efforts are Alexa Health, which is a trusted health assistant for patients and clinicians, Comprehend Medical, which is able to deal with Electronic medical Records, and various other initiatives. In this talk, we would like to address the following questions: • How different is to build a product for HCLS than any other domain ? • What are the biggest areas of interest for ML in HCLS ? • What are the challenges around data privacy and regulations when it comes to the adoption of ML use cases in HCLS? • How can companies build AI applications in HCLS and maintain customer trust? • What are some of the risks with data collection in HCLS ? • What is the importance of explainable models in HCLS?