Carta empowers hospitals to personalize the delivery of care to the individual needs of each patient. Our insight here is that personalizing care is not only good from an outcome/clinical perspective (where most people focus in the context of personalization) but is also the best way to optimize operations. Currently, hospitals have to over provision their resources because they are set up to serve the generic patient; planning ahead for exactly the resources needed- no more, no less- is the best way to gain efficiency. The approach we're advocating and enabling is to:1) Find past patients similar to the current one being treated2) Quantify what exactly happened to them during their journey through the hospital (this is where our model comes in)3) Use machine learning to project what the particular patient in question will need, and what the patient can expect their experience to be in the hospitalWe're applying this approach now to two use cases- supplies projection and bed usage projection- and we have a bunch of other use cases we're planning on addressing in the future.