NODE - Real-time data, AI, and the road to autonomous systems

All human-designed systems are on a path toward autonomy.

From wearable devices to industrial machinery, systems are evolving to adapt and self-correct, powered by high-resolution, instrumented data. This is time series data, and it’s the foundation of modern autonomy, enabling AI systems to make smarter decisions, predict outcomes, and optimize operations with minimal human intervention.

For the industrial sector, the march toward autonomy is unlocking new levels of precision and control. Time series data, combined with machine learning (ML) and predictive analytics, moves manufacturers from reactive decision-making to proactive optimization. This enables real-time monitoring, predictive maintenance, and seamless automation, turning every data point into a strategic lever for efficiency and innovation.

For AI, success depends on collecting large volumes of high-resolution data and using it intelligently. It’s not just about the quality of data, but also the quality and precision—factors that determine the effectiveness of AI systems.

High-resolution, time-stamped data is the fuel for AI/ML models. Sensors across industries—from smart manufacturing to satellites in space orbit—constantly collect this data, feeding it into machine learning models that learn, adapt, and improve. The result is AI systems that are not only smarter but also self-healing, self-adapting, and ultimately fully autonomous.

From data to intelligence

Every connected device, from simple apps to advanced industrial systems, generates a constant flow of time-stamped data. Whether recorded by the hour, minute, millisecond, or nanosecond, this data tells a story over time. While a single data point might seem insignificant on its own, it’s the larger patterns—the endless cascade of data over time—that allow AI systems to perform complex analytics, make predictions, and automate decisions. In turn, time series data enables industries to operate in real-time—ingesting, analyzing, and acting on data as it’s created.

Building the right AI architecture

To manage high-resolution data effectively, organizations need scalable infrastructure. Modern databases have made breakthroughs by supporting unlimited cardinality, enabling organizations to handle detailed, high-volume, high-resolution data without compromising performance.

Another central piece of infrastructure is the data lakehouse—a new paradigm combining the capabilities of data lakes and warehouses. It enables business intelligence and machine learning on all data, making it an ideal store for structured and unstructured time series telemetry data.

By collecting more high-resolution data, organizations can transform their data into actionable intelligence, creating more powerful systems; it’s these systems that will define the next wave of innovation

While data lakehouses provide the ideal AI datastore and can leverage pre-trained models, custom ML models may be necessary for unique data applications. Open file formats like Apache Parquet and Apache Iceberg ensure interoperability, enabling organizations to merge large datasets quickly.

AI is not a static technology, so monitoring and feedback are essential to its ongoing development. Tracking key metrics such as model response latency and accuracy ensures that AI systems remain optimized.

Delivering autonomous systems

AI-driven autonomous systems that require limited human oversight represent the highest level of AI. Enterprises investing in the ability to harness and process sophisticated telemetry data required to fuel these systems are creating significant competitive advantages.

Industry 3.0 introduced the digital revolution with programmable logic controllers and early automation. Industry 4.0 expanded this with IoT devices and real-time analytics. The next industrial revolution will harness high-resolution, time series data integrated with AI and ML. A continuous data stream enables systems to learn, adapt, and optimize in real-time, minimizing human intervention and maximizing efficiency.

Managing this data requires architecture capable of handling high ingest and query rates while scaling with long-term data growth. By collecting more high-resolution data, organizations can transform their data into actionable intelligence, creating more powerful systems; it’s these systems that will define the next wave of innovation. Organizations that transform their data into actionable intelligence will lead the way, creating smarter, more adaptive operations that define the future of human designed systems.

Evan Kaplan

Evan Kaplan is a seasoned entrepreneur and technology leader with over 25 years of executive experience. He is currently the CEO of InfluxData, the company behind InfluxDB, the leading time series database. Previously, Evan served as President and CEO of iPass Corporation, where he led its transformation into a global leader in Wi-Fi connectivity.