March 2022 Summaries

The Robust & Responsible AI Newsletter - Issue #1 provides an overview of the latest developments in MLOps and Data-Centric AI. Key highlights include Real-time Machine Learning by Chip Huyen, monitoring solutions for ML models presented by Loka, and the upcoming Data Council 2022 event in Austin. Open source spotlight features ZenML, whylogs, and Flyte, while MLOps experts are reading about data-centric AI, responsible AI, and academic research on monitoring ML and data systems. At WhyLabs, recent releases include the AI Observatory on AWS Marketplace, SOC 2 Type 2 certification, and new interactive features in the profile viewer. Upcoming events include a hands-on data monitoring workshop, MLOps World summit, and ODSC East conference.

WhyLabs, an AI observability platform, has announced its availability on the AWS Marketplace. This allows AWS customers worldwide to quickly deploy the WhyLabs AI Observatory for monitoring, understanding, and debugging their machine learning models deployed in AWS. The AI Observatory ensures users can be certain about the performance of their ML models. With this partnership, data scientists and machine learning engineers can start using the platform without talking to a salesperson or entering credit card information.

TeachableHub and WhyLabs are two platforms that make it easy to deploy models into production and maintain their performance. TeachableHub is a fully-managed platform for streamlining the deployment, serving, and sharing of impactful machine learning (ML) models as public or private APIs serverless with zero downtime. On the other hand, WhyLabs is an AI observability platform that enables users to achieve healthy models, fewer incidents, and happy customers. Together, these platforms provide a smooth monitoring and deployment solution for ML models. They integrate seamlessly, allowing users to log training and testing datasets, monitor model performance, and detect issues such as data drift. TeachableHub supports multiple environments by default, making it easy to create new ones and design custom processes for releasing new candidates to production. WhyLabs can be used to detect changes in real-world behaviors and highlight them through its dashboard. This helps users identify model performance degradation due to data drift and take appropriate actions such as retraining the model and redeploying it with TeachableHub. Overall, the combination of TeachableHub and WhyLabs streamlines the deployment process and automates repetitive tasks, enabling teams of all sizes to adopt established MLOps standards and best practices.

Observability is crucial for measuring the negative effects of technical debt in machine learning production systems. Many teams lack tools and processes to evaluate their ML models' ongoing performance after deployment, leading to potential issues going unnoticed until customers report them. Technical debt can be silent and insidious, causing catastrophic failures without triggering typical DevOps alarms on service and data availability. Observability into the dynamics of your data and models allows for proactive detection and response to changes in model performance before stakeholders or customers notice an issue. AI observability platforms like WhyLabs Observatory and open-source data logging libraries such as whylogs can provide purpose-built tools for large datasets, helping teams uncover the sources of ML technical debt effects and improve their models' performance.