Abstract: 

Many teams have data scientists and ML researchers who can build state-of-the-art models, but their process for building and deploying ML models is entirely manual. This is considered the basic level of maturity, or level 0, as defined by Google, and transitioning out of this state means addressing the technical debt around MLOps. Data scientists work hard to execute highly complex statistical models and analysis of datasets, but how do they report their work to the business and what is the fastest path to value? This demo works to demonstrate how a data scientist can take local environments, move it to a reproducible, scalable, and shareable compute environment, and hand off their results to the business and/or engineering personas. This standardization of hand-offs will help enable the creativity of the data science process by offloading the complexity ( and by extension cognitive load) of sharing data science work with organizations. A demo attendee will leave with an understanding of the potential complexity of sharing-like environments, where data science is more than JUST computer science, and how we can operationalize feedback loops within our MLOPs teams via centralized tooling.

Bio: 

Chase is a solutions architect at Arrikto with a passion for connecting people to technical solutions that can prevent them from wasting precious time and mental energy- solving the same problems over and over. Chase is a certified Kubernetes Administrator, Developer, and Security Specialist who works to help clients reduce MLOps friction and toil while ensuring the "non-negotiables" are enforced to provide the best return on their production models.