Avoca raises $125M+, reaches $1B valuation with AI agents for inbound calls and dispatch. Avoca raises $125M+ and reaches a $1B valuation by automating inbound calls and dispatch for HVAC, plumbing, and roofing businesses with AI agents. This shows agentic coordination driving real-world operations and dispatch workflows — a practical example of Orchestration and treating delivery as an organizational capability (Principle 09).

Inside OpenSearch’s bid to become the default AI data layer. OpenSearch 3.5–3.6 adds compressed vector search, neural sparse retrieval, and native agent memory APIs to position itself as a full AI data layer. Outcome engineers get a production-ready vector+memory substrate for agent memory, hybrid retrieval, and scalable embeddings — directly affecting how you design durable context and The Graph (Principles 06 & 11).

I built the Playwright for desktop apps — 80% token savings. agent-desktop exposes native accessibility trees to AI agents and uses skeleton traversal plus deterministic refs to cut token usage up to 80%. That materially reduces context costs and gives agents stable, programmatic handles into GUIs — a practical pattern for expanding agentic control beyond web UIs (Principles 06 & 11).

Flue is a TypeScript framework for building the next generation of agents. Flue provides a TypeScript agent harness and sandbox to build, run, and deploy autonomous agents anywhere. Treating a harness as a first-class framework standardizes runtime, sandboxing, and deployment paths so teams can ship agent artifacts and integrate agents into existing stacks (Principles 06 & 07).

The Agent Harness Belongs Outside the Sandbox. Andrea Luzzardi argues to move the agent harness off the sandbox to protect credentials, enable durable multi-user sessions, and treat sandboxes as disposable compute. That architectural shift changes where you place authority and secrets, enabling safer, multi-tenant agent services and better coordination across users and teams (Principles 07 & 03).