Speaker "Misha Kutsovsky" Details Back



Productive machine learning workflows in a hybrid cloud MLops world


Recent years have seen tremendous advances in state of the art machine learning models in domains such as speech, text & images. These deep learning models are extremely demanding on computational resources & require a new set of tools and processes for companies to take advantage of. Success depends on the scale of experimentation — running more experiments equals better results & teams need to be able to iterate across training these large demanding models and do so in a reproducible fashion. Misha is a PM at Paperspace working on Gradient, which is a serverless platform that makes it simple and fast to run machine learning and deep learning workloads of any scale and complexity on any infrastructure. He will outline the workflow of productive machine learning organizations so they can decrease time to market in implementing AI-enabled solutions across the enterprise.
Who is this presentation for?

Prerequisite knowledge:

What you'll learn?


Misha is a PM at Paperspace working on their hybrid cloud machine learning platform Gradient. Prior to this, he was at Microsoft's Windows Active Defense team building fileless malware detection software & on tooling that powered the end-to-end machine learning systems for Microsofts DevOps & Data Scientists teams. He has a M.S in Electrical & Computer Engineering from Carnegie Mellon.