Introduction: From AI Demos to Enterprise Operations

In 2026, the artificial intelligence landscape has definitively shifted from the laboratory to the factory floor. After years of deploying billions of dollars into capital-intensive foundation models, the venture capital ecosystem is aggressively rotating its focus toward the essential infrastructure required to make AI functional in the real world. The recent announcement that Orkes, an enterprise AI workflow orchestration platform, has secured a $60 million Series B funding round perfectly encapsulates this macroeconomic transition. As organizations move past the novelty of generative AI chatbots and demand concrete return on investment (ROI), the "unsexy plumbing" of the AI stack is emerging as the most lucrative segment of the tech industry.

The VC Pivot: Why Infrastructure is Winning

The first wave of the generative AI boom was defined by gargantuan investments in core model providers. Companies like OpenAI and Anthropic commanded massive mega-rounds, culminating in a reality where, in late 2025, a staggering 93% of venture capital dollars in Silicon Valley were flowing into AI. However, the economics of foundation models have proven challenging. Compute costs remain astronomically high, while API pricing has collapsed due to fierce competition. Many limited partners (LPs) who expected software-as-a-service (SaaS) margins are now reckoning with what effectively amounts to hardware company economics.

In contrast, infrastructure companies building the foundational layers of AI deployment represent a dramatically different investment proposition. Platforms that orchestrate AI workflows generate recurring revenue with gross margins frequently exceeding 80%—the exact unit economics that Series B investors demand. The overarching narrative has shifted: while foundational models provide the raw intelligence, it is the orchestration and infrastructure layers that build sustainable, defensive moats. During previous technology cycles, the companies that provided the "picks and shovels"—such as cloud infrastructure providers during the web boom—ultimately captured enormous and enduring value. The AI revolution is proving to be no different.

Orkes: The Netflix Pedigree

At the center of this infrastructure boom is Orkes, a Santa Clara-based startup founded in 2021 by the original engineering architects behind Netflix's microservices orchestration platform. CEO Jeu George, CTO Viren Baraiya, and CPO Dilip Lukose originally built the open-source "Conductor" project at Netflix in 2016 to manage the streaming giant's hypergrowth. Conductor was engineered to coordinate complex backend processes across microservices, scaling to handle billions of executions with fault tolerance and deep observability.

Orkes has spent the past four years adapting and commercializing this battle-tested foundation specifically for the age of "Agentic AI." While AI models can reason and generate text, they lack native mechanisms for state management, error handling, and system integration. Orkes bridges this gap by providing an enterprise-grade engine that coordinates Large Language Models (LLMs), microservices, internal APIs, and human review processes within a highly durable execution model.

Funding Details: Inside the $60M Series B

In April 2026, Orkes announced it had raised $60 million in fresh capital to accelerate the deployment of its platform. The Series B round was led by AVP (a global investment platform managing over €2.5 billion in assets), with AVP General Partner Alex Scherbakovsky joining the Orkes board of directors. The round also saw participation from new investor Prosperity7 Ventures, alongside existing backers Nexus Venture Partners, Battery Ventures, and Vertex Ventures US. This significant capital infusion follows a $20 million Series A round in 2024, bringing the company's total funding to over $80 million.

The investment reflects massive market traction. Orkes has successfully built a global community of hundreds of thousands of developers and recorded millions of platform installs. Over the last two years, the company has tripled its enterprise customer base, securing contracts with Fortune 100 leaders including JPMorgan Chase, Atlassian, Tesla, Oracle, American Express, and GE Healthcare. Notably, Netflix remains a core user of the underlying open-source technology, with internal usage of Conductor reportedly increasing by 5x in recent months as the streaming company scales its own internal AI initiatives.

Market Analysis: Closing the Execution Gap

The timing of Orkes' funding aligns with a critical bottleneck in the global tech economy. Gartner projects that AI software spending will reach $450 billion in 2026. Yet, according to a 2025 McKinsey report, a staggering two-thirds of companies were still stuck in "pilot mode." The reason is simple: an AI demo built in a sandbox breaks down almost immediately when exposed to the complexities of a live production environment.

Enterprise workflows demand absolute reliability. If an autonomous AI agent hallucinating an output triggers a financial transaction or alters a patient's medical record, the consequences are catastrophic. Organizations require an orchestration layer that enforces governance. This includes implementing retries for failed API calls, setting execution timeouts, mandating human-in-the-loop (HITL) approvals for sensitive actions, and maintaining full auditability of every decision the AI makes. Orkes positions itself as the missing execution layer that transforms fragile AI experiments into mission-critical, highly durable business operations.

Product Strategy and Implications

With the $60 million war chest, Orkes is aggressively expanding its suite of tools tailored specifically for AI developers. The company has introduced several pivotal features:

• Agent Runtime: A robust execution model that marries structured process stages with dynamic, LLM-driven decisions. This allows AI agents to operate within strict, reliable guardrails.
• MCP Gateway: A capability that instantly transforms an enterprise's legacy, internal APIs into secure, consistent tools that autonomous AI agents can interact with out of the box.
• Prompt-to-Workflow: A rapid development tool that allows engineering teams to convert natural language instructions into functional, editable baseline workflows, drastically accelerating time-to-market.

These strategic product rollouts indicate that Orkes is not just building a workflow engine, but rather an entire operating system for enterprise AI. By treating AI prompts and agentic behaviors as "first-class citizens" alongside traditional software microservices, Orkes is enabling organizations to modernize their legacy systems without requiring a complete, risky replatforming.

The Investor Perspective

From the viewpoint of venture capitalists, Orkes represents the ideal convergence of technical credibility and market necessity. Alex Scherbakovsky of AVP highlighted that developers have historically lacked the "enterprise-grade orchestration, observability, and security to move AI workloads into production safely." Orkes solves this immediate, expensive problem.

Furthermore, Abhishek Shukla of Prosperity7 Ventures emphasized that Orkes provides a "single, governed engine to coordinate LLMs, tools, microservices, and human review." For investors, this translates to extreme product stickiness. Once an enterprise integrates Orkes to run its core operational workflows, ripping it out becomes nearly impossible. This structural advantage, combined with the explosive growth of AI agent adoption, creates a highly defensible business model capable of sustaining long-term growth even if underlying foundational AI models become commoditized.

Conclusion: The Future Belongs to the Orchestrators

The $60 million Series B raised by Orkes is a clear signal that the AI hype cycle has matured. The industry is no longer simply marveling at the conversational abilities of large language models; it is demanding rigorous, scalable execution. As venture capital shifts away from the speculative margins of foundational models and toward the resilient economics of infrastructure, workflow orchestration platforms will become the most critical layer in the modern software stack. For founders, investors, and enterprise leaders alike, the lesson of 2026 is clear: the companies that build the smartest AI will capture the headlines, but the companies that build the infrastructure to control that AI will capture the market.