Partnership will advance monitoring capabilities in production AI systems
Mona, the intelligent monitoring for AI systems, announced a partnership with New Relic, the observability platform. New Relic is a powerful cloud-based observability platform that helps engineers instrument, analyze, troubleshoot and optimize their entire software stacks. Companies that currently use New Relic can now view dashboards with Mona insights that instantly detect anomalies about data integrity issues and model performance issues within their production AI systems, significantly reducing business risks. The partnership is focused on bridging the gap between data scientists and production teams, enabling fast and reliable Machine Learning Operations, such as data corrections, model retraining, or other mitigations.
As AI becomes a more central growth driver for many companies, it is crucial to monitor data and model behavior in order to avoid risks that can substantially impact businesses along with their customers. AI systems not only just include models, but they typically rely on multiple components (e.g. data pipelines, 3rd party data sources) which all can introduce integrity issues, format changes, and also other bugs. Moreover, the models themselves are usually very sensitive to the data and unexpected real-world changes can cause significant issues with model performance. Mona automatically analyzes data across the entire system to enable SREs and DevOps teams to have a complete understanding of the data performance. Without the proper monitoring capabilities, businesses may find it difficult to scale support and sustain quality business outcomes. Being able to bridge the gap between AI practitioners such as data scientists and production-oriented functions such as SREs, allows quicker resolutions to tackle performance issues.
Read More: Vistex Achieves POI Best-In-Class 2021: Foodservice, HQ Planning And Dashboards
Through this partnership, companies now have the ability to explore production machine learning data, created by Mona, directly within New Relic One platform. From there, dashboards containing Mona insights about their production AI system are automatically generated. The integration also enables users to create custom Mona alerts to New Relic Alerts and Applied Intelligence, a flexible and centralized notification system that automatically detects anomalies and reduces alert noise by correlating related alerts and incidents.
“We are thrilled that New Relic selected us as an MLOps integration partner”, said Yotam Oren, co-founder and CEO of Mona. “This collaboration validates the criticality of monitoring production AI systems, enabling our mutual customers to maximize insights and value from their data.”
Mona provides instant alerts on latency issues within production systems, notifying specific people with the proper alerts. With New Relic Alerts and Applied Intelligence, anomalies are automatically detected, alert noises are reduced through correlating related alerts, and also provide root cause analysis. This integration will allow businesses to detect, analyze and resolve data issues faster.
Read More: SalesTechStar Interview With Rachel Obstler, Head Of Product At Heap
“By partnering with Mona, New Relic delivers the only unified data observability platform that gives data science and DevOps teams unprecedented visibility into the performance of their machine learning models running in production” said Guy Fighel, General Manager of Applied Intelligence and Group Vice President of product engineering at New Relic. “Monitoring the effectiveness of the production models while enabling collaboration between Data Science and DevOps teams will make it easier to develop, test, and monitor sophisticated ML models to ensure more relevant, meaningful customer experiences and maximize business impact.”
With better visibility into AI systems, businesses can plan better AI roadmaps using the feedback based on production data. Now, New Relic allows production AI / ML data exploration via NRQL (New Relic’s query language). NRQL retrieves data in order to get insights into applications, hosts and important business activities. These additional monitoring capabilities enable teams to focus on the most important metrics for their company.
“The combination of New Relic’s platform and Mona’s insights take monitoring for AI systems to a whole new level”, said Itai Bar Sinai, co-founder and CPO of Mona. “Finally, there is a way for DevOps and SREs to rapidly resolve production issues that manifest in data and model performance.”