Cortex

The cloud platform behind our intelligent transit solutions.

Comprising compute resources, storage systems, data protection, orchestration tools and APIs, Cortex delivers our applications via a Software-as-a-Service (SaaS) model.

Making sense of transit data from many sources is complex, convoluted, and time-consuming. Streamriver Cortex – the cloud-native foundation behind our AI-driven transit solutions including Insights Analytics and TrainManager – unifies transit data, making it easier to understand and more immediately actionable. Built entirely on Amazon Web Services (AWS), Cortex ingests data from across your IT network and turns it into one secure, reliable source of operational intelligence.

Built to scale, designed to stay up

Cortex runs on containerized, serverless compute that scales with demand, so a regional operator and a national network get the same performance. Data is stored in a managed cloud database with multi-availability-zone replication, automatic failover and encryption at rest. Application services run in isolated private networks, and all traffic is encrypted in transit.

real-time and batch, in one platform

Cortex handles live transit data streams such as train telemetry, vehicle positions and on-board device status alongside scheduled processing for nightly ingestion, KPI rollups and automated reports delivered by email. ML models review every journey daily to detect anomalies.

open by design

Streamriver supports ITxPT fundamental concepts for open architecture, data accessibility and interoperability between IT systems. Cortex integrates with a wide range of transit and enterprise data sources, including EMV contactless payment platforms, onboard Wi-Fi and cellular communication providers, GNSS and live train feeds, GTFS schedules, service desk and CRM tools, and more. Systems on private networks connect securely, and field devices such as fare-inspection terminals use mutually authenticated TLS.

AI built in, not bolted on

Foundation models hosted within AWS power the Streamriver AI Assistant, letting staff ask questions of their data in plain language, along with automated report narratives and anomaly triage. A built-in knowledge base answers questions from your own documentation, ingested directly from PDFs and other formats. Cortex also supports the Model Context Protocol (MCP), the open standard that defines how AI agents connect to external tools and data sources. Via the Insights MCP server, authorized users can securely query their transit data using external AI apps and agents such as Anthropic Claude and OpenAI ChatGPT.

Secure and fully monitored

Users sign in through your organization's identity provider via OpenID Connect, with permissions enforced per application. Every release passes through automated builds and a staging environment before a manual approval gate for production. End-to-end tracing, synthetic monitoring and real-user performance tracking provide continuous visibility, with alerts routed to on-call engineers around the clock.

One foundation, every solution

Because every Streamriver solution runs on Cortex, your applications share one trusted data layer and strengthen each other. Consist data from TrainManager, for example, tells Insights Analytics exactly which passenger cars are running on which trains, so Wi-Fi, backhaul and contactless payment system performance can be traced to specific vehicles and routes. New solutions plug straight into the data and integrations you already have, with no new silos and no duplicated effort.

Cortex drives our application portfolio. Contact us today to discuss your needs and arrange a live demonstration.