Selector granted 8 U.S. patents for AI-powered network intelligence

Selector, a leader in AI-powered network operations intelligence, announced that the United States Patent and Trademark Office (USPTO) have granted 8 foundational patents to the company.

According to the company, this marks “a significant milestone in Selector’s mission to advance the science of causal reasoning, natural language interaction, and predictive analytics within complex digital infrastructures.”

The granted patents span innovations in causal inference, large language model (LLM) training, AI-powered correlation, predictive maintenance, and network path intelligence, reinforcing Selector’s position at the forefront of next-generation AIOps and observability technology.

Each invention reflects the company’s deep expertise in applying machine intelligence to the data, events, and dependencies that shape network behavior, the company said.

The granted patents include:

• Root Causation for Network Operations — Introduces AI-driven causal reasoning to pinpoint fault origins across multi-domain environments, dramatically reducing Mean Time to Resolution (MTTR).
• Dashboard Metadata as Training Data for Natural Language Querying — Uses visualization and interaction metadata to train Selector’s network-specific large language model (LLM), enabling more intuitive and context-aware natural language queries.
• Metrics, Events, Alert Extractions from System Logs — Transforms unstructured telemetry into structured, correlated insights, enabling consistent analytics and faster anomaly detection.
• Methods and Apparatus for Network Tracing, Forecasting, and Capacity Planning — Applies advanced analytics and modeling to predict capacity risks and network growth before performance is impacted.
• Methods and Apparatus for Determining a Path that a Data Packet Would Traverse Through a Communication Network at a Time of Interest — Enables precise reconstruction of packet paths through a network at a specific point in time, improving historical analysis, forensic investigation, and root cause accuracy.
• Early Identification of Optical Transceiver Failures — Uses predictive modeling to spot hardware degradation early, allowing teams to replace failing optics before they cause an outage.
• Methods and Apparatus for Efficient Storage and Querying of Communication Network Parameters — Introduces scalable techniques for storing, indexing, and querying large-scale network topology and routing state.
• Maintenance Window Aware Reporting — Automates the detection and exclusion of maintenance windows from performance analytics, improving the precision of service availability and reliability metrics.

“These patents reflect years of focused innovation to bring AI and causal reasoning to the heart of network operations,” said Nitin Kumar, CTO and co-founder of Selector. “Selector’s platform doesn’t just monitor data, but actually understands relationships, predicts failures, and explains why events occur. These innovations are foundational to how we’re reimagining observability for the AI era.”

Each patent reinforces a core element of the Selector platform—from its correlation engine to its network-trained LLM—creating a unified framework for understanding cause and effect across distributed systems.

For more information about this news, visit www.selector.ai.