The recent partnership between Aporia, the customizable ML observability platform, and ClearML, the unified, end-to-end MLOPs platform, has prompted the release of their latest ML pipeline optimization solution. Designed to target time-to-value and time-to-revenue margins, this collaborative solution employs a holistic quality assurance of ML projects to advance its commercial production time, according to the companies.
Aporia and ClearML emphasize their drive towards aiding data scientists and ML engineers; often left to operate with disparate tools and a myriad of moving parts within ML projects, these data experts undertake significant workloads that leave an opportunity for technological innovation. The combined efforts of Aporia and ClearML seek to remedy these challenges with a frictionless, end-to-end integrated solution that optimizes workflow throughout the ML project, from production, deployment, and monitoring, according to the companies.
"As an open source company dedicated to giving the data science, ML engineering, and DevOps communities the tools they need to do more with their machine and deep learning projects, we're excited to integrate with Aporia to add cloud-native ML observability to our unified, end-to-end MLOPs platform," said Moses Guttman, CEO and co-founder of ClearML. "The result of this joint effort means that our customers can do even more from the very start without the friction—using ClearML to build, train, orchestrate, and serve their models seamlessly and with just two lines of code; and using Aporia to monitor, explain, and improve those models once they hit production.”
ClearML’s ML streamlining technology works in conjunction with Aporia’s customizable ML observability solution, integrating CI/CD automation approaches from ClearML with Aporia’s explainable AI. Through its automation approach, ClearML ensures ML development and production work is reproducible and scalable, ultimately reducing time-to-value and time-to-revenue by optimizing developer workflow.
As soon as the model is deployed into production by ClearML, Aporia enables users to monitor, explain, investigate, and solve issues involved with their AI, fostering trust. Users can analyze issues such as data and concept drift, performance degradation, and model decay; when a model is spiraling, customizable model monitoring triggers live alerts and generates a comprehensive visibility dashboard that provides an “Investigate and Explain” function for further guidance post-deployment.
“As an MLOps leader ensuring data science and ML teams can trust their ML model predictions, we see immense value for our customers in integrating with ClearML’s open source MLOps platform to provide a true end-to-end solution from training to production and beyond,” said Liran Hason, CEO and co-founder of Aporia. “There are so many different tools and moving parts to pull from when setting off on this hero’s path to build, train, serve, monitor, and explain machine learning models. We’re excited to team up with ClearML and provide a one-stop-shop MLOps platform to scale ML with confidence.”
For more information about the partnership and its joint solution, please visit https://www.aporia.com/ or https://clear.ml/.