.png)
Since our Multi-Agent Orchestrator launch in December 2024, we’ve introduced new features and refined capabilities to streamline operations, strengthen compliance, and support evolving enterprise needs. Below are the highlights.
Chat Interface
A new chat interface is now available (multi-turn conversations coming soon!). It simplifies user registration, onboarding, and usage tracking, while providing a direct channel to test or demonstrate how agents respond to targeted prompts. Developers and users can quickly try out agent functionality without configuring APIs.
The platform integrates with any service that exposes an OpenAPI spec. Out-of-the-box connectors for Jira, Confluence, and Amadeus (Travel API) are now included. Jira and Confluence connectors are available on our Enterprise tier (contact us for more). Amadeus is part of our Public tier.
The API Registry supports 100+ APIs and can handle specification conversions as needed. A dedicated Connector Agent automatically determines which API calls to make and manages credentials in a secure environment. See demos of how the Orchestrator uses the API Connector with Jira, Confluence and Amadeus Travel.
Integration with CrewAI and LangChain
Emergence now has integration with CrewAI and LangChain. These frameworks help developers build , multi-step agent workflows and chain together large language models for advanced reasoning. By calling the Emergence Orchestrator via its API, these frameworks can trigger tasks through our Agent SDK.
Get started with example projects and demo videos for CrewAI and LangChain:
Browser Live View
Users can now watch web-based automation in real time. Browser Live View displays step-by-step actions—page navigation, form filling, or other interactions—so teams can verify workflows and adjust them on the fly. This immediate visual feedback reduces errors and speeds up fine-tuning. Checkout our docs and demo for details.
Workflow Visibility and Trust
Detailed logs and real-time diagnostics enhance transparency for both developers and operational teams. This level of visibility builds confidence in automated workflows, simplifies troubleshooting, and maintains consistent performance across large deployments. See de
We’ve deployed a new, more robust model for safety checks at the platform level. Disallowed or suspicious content is flagged before it reaches any agent, and enterprises can customize these policies to align with internal compliance requirements. Additionally, the API can now be integrated with broader governance and compliance workflows, further centralizing oversight of AI-driven processes.
These updates set the stage for a secure, transparent, and highly scalable multi-agent orchestration platform. More enterprise features are on the way. Stay tuned, and happy building!
A first-principles architecture for agentic systems, inspired by Unix and Kubernetes. It introduces nine core abstractions—such as Execution Contexts, Skills (as high-level system calls), and Dynamic Agent Instantiation—to enable runtime agent creation, recursive delegation, and asynchronous execution.
A structured five-layer framework provides standardized benchmarking for AI agent capabilities across the full spectrum of enterprise task complexity, from UI to infrastructure.
Discover why standard AI benchmarks fall short for enterprise needs and how agent performance is truly measured on realistic, multi-application workflows using both UI and APIs.