Engineering Blog
Technical articles, updates, and insights on building production AI agent infrastructure.
Building in Public: The v0.8 to v0.9.1 Chapter
What changed between v0.7 and v0.9.1 — Planner orchestration, Tracker observability, the schema Registry, and a two-layer SDK that keeps agent code clean. v0.9.1 is available now.
Service Discovery and the Schema Registry in Soorma
How Soorma's Registry service enables agents to find each other by capability, advertise their event contracts, and form a discoverable, self-documenting agent ecosystem.
Tracker Observability: Understanding Plan and Task State in Soorma
How the Soorma Tracker service gives you full visibility into multi-agent plan execution — from task emission to terminal state — without instrumenting your agent code.
Inside the Planner: Event Choreography and Strategic Orchestration in Soorma
A technical deep-dive into how the Planner component uses explicit response_event, plan state machines, and the DisCo pattern to orchestrate multi-agent workflows.
Soorma Core v0.7: The 'No-Magic' Guide to Event-Driven Agents
Building in public: Soorma Core v0.6 and v0.7 are ready for learning and prototyping. We want your feedback—do the learning paths help? Does the DisCo framework make sense?
Zero to AI Agent in 10 Minutes: The Low-Code Path to Building Intelligent Systems
How Soorma brings the simplicity of Firebase to AI agents—no complex setup, no infrastructure headaches, just code.
It's Time to DisCo: The Case for Distributed Cognition in AI Agents
Why we need to move beyond single-threaded loops and embrace event-driven cognitive architectures.