Seldon’s Marcus Hinterseer will outline the common challenges organisations face when operationalising machine learning, from model drift to compliance and risk challenges, as well as the key techniques to overcome them. Deploying these complex models can tie up development effort for months and, if badly managed, pose seriousSpea risk to organisations. Marcus will dive into how to approach deployment responsibly and efficiently through powerful monitoring, management and explainability techniques.
Key Takeaways
- What are common challenges organisations face when creating a ML deployment pipeline?
- How can MLOps techniques bridge the gap between Data Science and DevOps teams?
- What infrastructure do organisations need to have implemented at different stages of the ML Maturity model?
- How can you bake in monitoring and other risk-minimising methods into your deployment pipeline?
Add comment