Key Takeaways
- • Poetic, founded by former Google and Waymo ML engineer Markie Wagner, closed a $50 million Series A at a $500 million valuation, led by Kleiner Perkins with participation from OpenAI, Founders Fund, and First Harmonic.
- • The company achieved an eight-figure run rate in 2025 with only four employees and posted a 100% pilot-to-production conversion rate at marquee clients SoFi and AIG, demonstrating significant enterprise traction.
- • Poetic's proprietary technology departs from autonomous agent architecture by deploying a domain-specific programming language that allows operations teams to describe complex workflows in natural language, designed to deliver production-grade reliability in high-stakes enterprise processes.
- • Wagner's founding thesis emerged from direct observation of AI system failures in production environments across two of the most advanced AI organizations—Google and Waymo—where models performed in controlled settings but collapsed under real operational complexity.
- • The investment reflects institutional validation that enterprise AI deployment requires architectural redesign rather than model improvement, positioning Poetic to address one of the sector's most persistent execution gaps.
Summary
Poetic has emerged from stealth with a $50 million Series A funding round at a $500 million valuation, led by venture capital firm Kleiner Perkins. OpenAI participated alongside Founders Fund and First Harmonic, signaling strong conviction among leading AI investors in the startup's approach to enterprise artificial intelligence deployment.
Founder Markie Wagner brings credibility from two premier AI organizations. As an ML engineer at Google and subsequently at Waymo, she developed an operational perspective on a critical problem: AI systems that demonstrate impressive performance in controlled environments consistently fail when deployed into real enterprise workflows. These failures stem not from model inadequacy but from architectural unsuitability for high-stakes, multi-step processes where error rates below a critical threshold are non-negotiable. After her tenure at both companies, Wagner founded Delphi Labs, an ML consultancy deploying AI for major enterprises, where she continued observing the same failure pattern across Fortune 500 implementations.
Rather than attempting to improve autonomous reasoning systems, Poetic adopted a fundamentally different architectural approach. The company built a proprietary domain-specific programming language that enables operations teams to describe their most complex processes in natural language. This methodology prioritizes deterministic execution over autonomous reasoning, functioning more like code than like a reasoning agent. The design philosophy directly addresses the production reliability gap that has plagued enterprise AI adoption.
Poetic's early commercial traction validates the market opportunity. Operating with only four employees in 2025, the company achieved an eight-figure annual run rate while maintaining a 100% conversion rate from pilot programs to production deployment at sophisticated financial services and insurance clients, including SoFi and AIG. This metrics profile—minimal headcount paired with substantial revenue and near-perfect client retention through deployment—indicates both product-market fit and operational efficiency that justifies institutional backing at scale.
The funding round underscores a strategic recognition within leading venture and AI circles: enterprise AI success depends on rearchitecting how AI systems integrate into workflows, not merely on advancing model capabilities. OpenAI's participation alongside traditional venture capital suggests the AI capability provider itself views production-grade enterprise deployment as a distinct layer requiring specialized infrastructure and methodology.
Founder Markie Wagner brings credibility from two premier AI organizations. As an ML engineer at Google and subsequently at Waymo, she developed an operational perspective on a critical problem: AI systems that demonstrate impressive performance in controlled environments consistently fail when deployed into real enterprise workflows. These failures stem not from model inadequacy but from architectural unsuitability for high-stakes, multi-step processes where error rates below a critical threshold are non-negotiable. After her tenure at both companies, Wagner founded Delphi Labs, an ML consultancy deploying AI for major enterprises, where she continued observing the same failure pattern across Fortune 500 implementations.
Rather than attempting to improve autonomous reasoning systems, Poetic adopted a fundamentally different architectural approach. The company built a proprietary domain-specific programming language that enables operations teams to describe their most complex processes in natural language. This methodology prioritizes deterministic execution over autonomous reasoning, functioning more like code than like a reasoning agent. The design philosophy directly addresses the production reliability gap that has plagued enterprise AI adoption.
Poetic's early commercial traction validates the market opportunity. Operating with only four employees in 2025, the company achieved an eight-figure annual run rate while maintaining a 100% conversion rate from pilot programs to production deployment at sophisticated financial services and insurance clients, including SoFi and AIG. This metrics profile—minimal headcount paired with substantial revenue and near-perfect client retention through deployment—indicates both product-market fit and operational efficiency that justifies institutional backing at scale.
The funding round underscores a strategic recognition within leading venture and AI circles: enterprise AI success depends on rearchitecting how AI systems integrate into workflows, not merely on advancing model capabilities. OpenAI's participation alongside traditional venture capital suggests the AI capability provider itself views production-grade enterprise deployment as a distinct layer requiring specialized infrastructure and methodology.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.