{"version":1,"type":"story","url":"https://digestai.news/story/joni-s-approach-to-ai-agent-orchestration","json":"https://digestai.news/story/joni-s-approach-to-ai-agent-orchestration.json","markdown":"https://digestai.news/story/joni-s-approach-to-ai-agent-orchestration.md","slug":"joni-s-approach-to-ai-agent-orchestration","headline":"JONI's Approach to AI Agent Orchestration","summary":"JONI, an Israeli platform, addresses the challenges of AI agent orchestration by providing persistent runtimes and execution capabilities. Unlike traditional models that focus on content generation, JONI offers a comprehensive layer above foundation models for executing tasks. The company allocates each user a persistent cloud runtime to hold memory, files, integrations, and scheduled tasks, ensuring background work continues between sessions. Compute-intensive operations are provisioned on-demand as ephemeral instances, while heavy processing runs in isolated sandboxes. JONI's architecture includes model access through a gateway abstraction, allowing substitution between providers without application changes. The platform routes requests to the best-suited connected models and completes actions rather than terminating at output generation. JONI supports various tasks such as domain registration, hosting deployment, advertising campaigns, media generation, telephony, and email operations. The company also offers a marketplace for third-party developers to publish agents and skills into user environments, with revenue shared in the publisher's favour. As of 2026, the agentic AI market is estimated at around $9 billion, growing to between $35 billion and $45 billion by 2030, indicating a significant opportunity for JONI.","keyPoints":["JONI provides persistent runtimes for each user","Heavy processing runs in isolated sandboxes","Supports various tasks including domain registration and hosting deployment"],"whyItMatters":"JONI's approach to AI agent orchestration is significant as it addresses the challenges of multi-step reliability, context persistence, and human oversight in AI systems. This can lead to more reliable and scalable AI applications.","category":{"slug":"enterprise","name":"Enterprise & Industry","url":"https://digestai.news/category/enterprise"},"entities":{"companies":["Mezada Development and Software Ltd.","Portkey","Langdock","Kore.ai"],"models":[],"people":[]},"firstPublishedAt":"2026-09-15T17:00:24Z","updatedAt":"2026-09-15T17:00:24Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"KDnuggets","title":"Orchestration and Execution: How JONI Approaches the Agent Layer","url":"https://kdnuggets.com/2026/09/finpr/orchestration-and-execution-how-joni-approaches-the-agent-layer","publishedAt":"2026-09-15T17:00:24Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"JONI's Approach to AI Agent Orchestration\", 15 September 2026, https://digestai.news/story/joni-s-approach-to-ai-agent-orchestration","publisher":"Digest AI","title":"JONI's Approach to AI Agent Orchestration","datePublished":"2026-09-15T17:00:24Z","url":"https://digestai.news/story/joni-s-approach-to-ai-agent-orchestration"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}