Juniper Networks announced several new enhancements that make it even easier to deliver predictable, reliable and measurable user experiences from client to cloud.

By integrating ChatGPT with Marvis, a virtual network assistant (VNA) driven by Mist AI, Juniper customers and partners can now easily access public-facing knowledgebase information using ground-breaking Large Language Models (LLM).

In addition, new Marvis integrations with Zoom enable superior video conferencing experiences while significantly reducing troubleshooting costs. With these enhancements, plus a new Wi-Fi 6E access point, Juniper is expanding its AIOps offering.

“AI is the next step in automating tasks that typically require a human IT domain expert, improving how IT teams operate the network with AI-driven tools like Marvis and its conversational interface,” said Bob Friday, Chief AI Officer at Juniper Networks. “Juniper Mist has always been a pioneer in utilizing proven AIOps to deliver assured user experiences from client to cloud, and with these latest LLM enhancements, Marvis will provide even more actionable knowledge and be an even more valuable member of the IT team.”

The Marvis VNA and its conversational interface were first introduced on June 7, 2018, as an essential part of IT, delivering proactive troubleshooting, predictive actions and exceptional insight into user experience via natural language processing and understanding (NLP/NLU). This enabled Juniper customers to easily delve into the network, user and application experiences (in real time) via simple language queries.

With the recent launch of LLM tools like ChatGPT, Juniper has been able to expand the conversational interface (CI) capabilities of Marvis to deliver even more human-like conversational capabilities, particularly with respect to documentation and support issues. Specifically, Marvis now leverages a LLM API to respond to user queries for technical documentation and other publicly available historical knowledge base information. For example, customers can ask Marvis “What do the Access Point LED lights mean?” or “List steps to configure Juniper campus fabric” and receive an accurate and direct response in the typical ChatGPT style in addition to a list of relevant documents.

Juniper is also leveraging 3rd party user-experience data from the Zoom cloud. By joining gigabytes of user experience Zoom data with gigabytes of network feature data, Marvis now has a deep learning model that can accurately predict user experience performance, allowing Marvis to use advanced AI/ML explainability techniques to quickly identify the root cause of video conferencing problems. In addition to real-time proactive troubleshooting (and self-driving corrective actions, if possible), Marvis learns trends to quickly detect and correct anomalies as well as predict future issues. This insight gives IT teams an edge in reducing Zoom support tickets and the time to repair issues.

Users can now leverage the Marvis conversational interface to easily access Zoom data insights via simple language queries, like “What was wrong with John Smith’s Zoom call?” or “List users with a bad Zoom experience.”