Real Time Sentinel: An LLM Based PII Detector -- A Streaming Integration and Intelligence Platform

Today's data-driven AI applications need to process, analyze and act on real-time data as it arrives, not wait for it to traverse a slow and complex pipeline. The massive amount of data is continuously generated from multiple sources and they come in a streaming fashion with high volume and high velocity, which makes it hard to process and analyze in real time. We introduce novel Gen AI based enhancements to Striim, a distributed streaming platform that enables unified real-time data integration and intelligence. Striim provides high-throughput, low-latency event processing with a scale- out architecture. It can ingest streaming data from multiple sources, process data with a SQL-like continuous query language, analyze data with sophisticated machine learning and large language models, write data into a variety of targets, and visualize data to support real-time decision making. It guarantees Exactly-Once-Processing semantics and fault tolerance. In this demonstration, we showcase Striim's ability to identify sensitive data over fast-moving data streams in real time using LLMs.