SANTA CLARA, Calif. – Orkes, the agents and durable workflow orchestration platform built by the original architects of Netflix’s microservices orchestration platform, has raised $60 million in fresh capital as adoption of its platform accelerates among developers building and deploying AI applications in production. AVP led the Series B round for Orkes, with participation from new investor Prosperity7 Ventures, alongside existing investors Nexus Venture Partners, Battery Ventures, and Vertex Ventures US.
Relied on by Developers, Trusted by Enterprises
Since raising its $20 million Series A in 2024, Orkes has built a global community of hundreds of thousands of developers, with millions of installs, and has tripled its customer base, including one of the world’s largest global retailers, United Wholesale Mortgage, Quest Diagnostics, Twilio, LinkedIn, European end-to-end commerce giant Naveo Commerce, and Australian global energy company Woodside Energy.
“Orkes’ platform brings a richness of engineering capability, and ease at which we are able to outsource state machine processing. To complement a great platform, we’ve been really spoilt with Orkes’ team helping us make the re-platform/transition seamless,” said Shaun Clements, principal engineer at Woodside Energy.
In addition, developers at thousands of organizations — including Netflix and other Fortune 100 leaders such as JP Morgan Chase, Atlassian, Tesla, Oracle, American Express, and GE Healthcare — continue to rely on the open-source Conductor project, originated while the Orkes founders led engineering efforts at Netflix in 2016 and now actively maintained by Orkes. Internally, Netflix’s usage of Conductor has increased by 5x in the last few months.
From AI Prototype to Production Deployments: Closing the Agent Execution Gap
2026 marks a pivotal year for AI, as enterprises shift from experimentation to proving ROI. Gartner projects AI software spending will reach $450 billion in 2026, yet McKinsey found that two-thirds of companies were still in pilot mode in 2025. Developers have lacked the enterprise-grade orchestration, observability, and security to move AI workloads into production safely.
“Orkes gives developers the confidence to deploy AI in production with the reliability, control, and scale they expect, and global enterprises rely on the platform for mission-critical operations,” said Alex Scherbakovsky, general partner at AVP who led Orkes’ Series B round and is joining the board. “That combination of deep technical credibility and real-world adoption is exactly what we look for at AVP, and we’re excited to support Orkes as they define how AI is operationalized in the enterprise.”
Orkes’ underlying open-source technology, Conductor, was originally built at Netflix to accelerate its hypergrowth to global scale, and continues to support Netflix today. Orkes has spent the past four years extending that battle-tested foundation to AI and agentic systems. As a result, developers can run agents confidently in production through Orkes’ execution layer. Unlike AI demos that break down quickly in real-world production environments, Orkes agentic workflows have strong built-in governance and reliability, high observability into AI actions, and can scale to high volumes easily with Conductor as their runtime.
Today, developers use Orkes to build, orchestrate, and operate AI agents and mission-critical workflows in production — with the reliability, guardrails, and observability that enterprise environments demand.
- Agentspan. Agentspan is an open-source durable runtime for AI agents. Agents built on Agentspan execute on Conductor’s proven orchestration engine, giving them the same durability, observability, and scale that powers workflows at Netflix and thousands of enterprises today.
- Agentic Workflows. Blend structured process stages with LLM-driven decisions, tool usage and human-in-the-loop capabilities in a highly durable execution model that is designed to handle real-world complexity.
- MCP Gateway. Turn internal APIs into safe, consistent tools that agents and LLMs can use right away.
- Prompt-to-Workflow. Turn natural language into a strong starter workflow you can review, edit, and ship fast.
Since Orkes launched its first agentic orchestration tools in 2023, developers have successfully deployed AI in production and it has become the fastest adopted capability within the Orkes platform. Naveo Commerce, for example, uses Orkes to power dynamic fulfillment and adaptive orchestration across a volatile global supply chain, enabling AI agents to autonomously monitor inventory, detect disruptions, and resolve issues in real time.
“Commerce is entering a new era. At Naveo, we see agentic orchestration not as incremental innovation but as a structural leap forward — enabling enterprises to adapt in real time, unlock new revenue models, and achieve resilience at scale,” says Naveo Commerce CEO Jamie Goldring. “Our partnership with Orkes will redefine how order and warehouse management is delivered for the next decade.”
“Developers need orchestration, controls, and visibility to run advanced AI and agentic systems with confidence. Orkes provides that assurance so developers’ applications and agents behave predictably at scale, even as they push the boundaries of what AI can do,” said Jeu George, co-founder and CEO of Orkes. “The trust we’re seeing from both the developer community and global enterprises reinforces the role Orkes plays as the missing orchestration layer that makes AI highly reliable, observable, and governable.”
“Orkes is helping define what production-grade AI orchestration looks like,” said Abhishek Shukla, Managing Director at Prosperity7 Ventures US. “Its platform gives enterprises a single, governed engine to coordinate LLMs, tools, microservices, and human review, so AI can safely sit in the middle of mission-critical workflows. We’re excited to support the team as they bring this capability to more customers and industries.”
