Logz.io uses generative AI to help operations teams solve problems

Cloud observable platform logz.io today announced it is introducing a generative AI model to augment its existing machine learning tools, which focus on an organization’s observable data, with additional insights from other data sources. As a result, the tool can now generate smarter remediation recommendations and, in turn, improve response times for operations teams.

The company, which focuses on collecting data from various open-source observation tools, started using supervised machine learning in 2016 and later expanded to include Cognitive Insights, based on crowdsourced insights from Logz. .io Community and other thread-based social networks. forums and open source repositories.

“Previously, Cognitive Insights generated crowdsourced recommendations,” said Asaf Yigal, CTO of Logz.io. “There is a good chance that there is useful and important information among those billions of events, and the system is trying to determine which links or discussion forums can provide additional context to solve the problem. But with the use of ChatGPT, we’ve added the ability to make more accurate recommendations based on crowdsourcing, offer possible solutions, and other ways to investigate the issue. In short, we can now give more accurate advice.”

Yigal stressed that Logz.io has a very unique set of data, including its users’ search behavior and metadata. In combination with OpenAI models, the service can now generate detailed and contextual remediation recommendations.

“Unlike other ML technologies, the main difference is access to contextual data about how engineers investigate problems across millions of libraries and software products,” he said. “That access to proprietary data is what makes this new system so fundamentally different. The more people use it, the smarter the system becomes. General answers are useless, while specific contextual advice can be groundbreaking.”