Managing and scaling ML workloads have never been a bigger challenge in the past. Data scientists are looking for collaboration, building, training, and re-iterating thousands of AI experiments. On the flip side ML engineers are looking for distributed training, artifact management, and automated deployment for high performance, scale, and cost efficiency. To add to the challenge there is no one size fits all solution and solutions are heavily dependent on the team size, AI domain, resource budgets, security, and cloud platform.
This session focuses on the core principles and design considerations while designing an end-to-end ML Ops pipeline. Different designs for version control, pipelines, training, and deployment will be discussed for both engineers and data scientists based on where they stand in their MLOps journey.
