Solo.io Launches Agent Substrate and kagent for Enterprise
Cambridge, Massachusetts-based Solo.io announced the general availability of enterprise releases of Agent Substrate and kagent on November 10, 2026, at 10:00 a.m. Eastern Time (GlobeNewswire). The company said the two products were "generally available today," marking the shift of its agentic infrastructure stack from preview to production-ready enterprise software (GlobeNewswire).
What's shipping
Solo.io describes Agent Substrate as an infrastructure layer purpose-built for agentic runtimes, combining serverless scale, sandbox isolation, and governed connectivity so enterprises can run AI agents without building that plumbing themselves (Solo.io). kagent sits alongside it as developer tooling for building and managing agents, tools, and skills — the kind of workbench developers need to assemble agent behaviors rather than just host them. Together, the pairing positions Solo.io to treat agents the way earlier infrastructure vendors treated containers: as a workload class that needs its own runtime, isolation, and connectivity guarantees baked in from the start.
Customers already in production
Solo.io named OutSystems and Swissquote as launch customers already running the infrastructure in live production environments, not pilot or sandbox deployments (GlobeNewswire). The most concrete deployment detail comes from Swissquote, where financial-crime investigation agents and associated MCP servers — internally called Codename Diego — are already running in production on kagent's general-availability release. The statement was attributed to Andrew Boswell, Head of Information Technology at Swissquote (Solo.io).
That detail matters because it shows the infrastructure isn't just handling experimental chatbots — it's underpinning agents tasked with compliance-sensitive work inside a regulated financial institution, where sandbox isolation and governed connectivity are not optional nice-to-haves but operational requirements.
Why it matters for the agent stack
The GA release lands as enterprises increasingly look for infrastructure that treats autonomous agents as first-class workloads rather than add-ons to existing application servers. Solo.io's pitch — serverless scale plus sandboxing plus governed connectivity, wrapped in developer tooling via kagent — targets exactly the gap companies hit when moving agents from demo to production: how to isolate what an agent can touch, scale it elastically, and give developers a consistent way to define its tools and skills. With named production customers in fintech and enterprise software already live, Solo.io is using concrete deployment evidence rather than roadmap promises to make its case in a crowded infrastructure market covered extensively around events like AgntCon and KubeCon (Solo.io).


