Labelbox Secures $100M Series E to Scale AI Agents and RL
Labelbox reaches $1.25 billion valuation to advance reinforcement learning and managed agent infrastructure.
1Series E Funding Milestone
Labelbox announced a Series E financing round exceeding $100 million on October 14, 2026. Led by Infinitum Partners, the deal values the company at $1.25 billion pre-money. The capital combines equity with a credit facility, specifically intended to bolster research and engineering for its core Horizon and Recursion Managed Agents product lines.
2Strategic Investor Alignment
The round included participation from Wipro Ventures, Andreessen Horowitz, B Capital, Tau Ventures, and Snowpoint Ventures. CEO Manu Sharma highlighted the importance of compute access, noting that Infinitum’s John Yetimoglu, an early Groq investor, provides strategic value. This partnership aims to support the company’s infrastructure needs as it scales its AI agent operations.
3Rapid Growth and Profitability
Labelbox achieved profitability in 2026 while reporting a fivefold year-over-year growth rate. The company currently serves over 400 customers, including more than 90% of leading U.S. AI labs. Its diverse client base spans hyperscalers, startups, and Fortune 500 enterprises, validating the platform's utility across various high-stakes AI research and development environments.
4Evolution of Product Strategy
Founded in 2018, the company evolved through three distinct chapters: initial training-data software, post-ChatGPT reinforcement-learning environments, and current agentic systems. Sharma emphasizes that Labelbox builds every platform layer internally, including sandbox infrastructure for millions of virtual machines and proprietary reinforcement-learning post-training systems, supported by an expert network of four million people.
5Horizon Simulation Capabilities
The Horizon product utilizes WorldSim to create simulated companies filled with realistic business data. Models attempt multi-step tasks while deterministic verifiers score trajectories. When two trajectories succeed, expert preferences teach the model specific taste. This system operates at the scale of thousands of worlds and millions of graded trajectories to meet pass@k targets.
6Recursion Managed Agents Launch
Launched in October 2026, Recursion Managed Agents execute goals based on triggers like Slack messages or API calls. A separate grader agent evaluates work against a specific rubric, revising tasks when they miss requirements. Agents operate in multi-agent, team, or auto modes, ensuring flexibility across over 100 business applications and MCP servers.
7Security and Compliance Standards
Labelbox prioritizes security by scoping tools as read-only or destructive, with credentials held in a secure vault. Sessions run in isolated sandboxes with zero data retention at LLM providers. The platform maintains compliance with SOC 2 Type II, GDPR, CCPA, NIST 800-171, and PCI DSS, ensuring enterprise-grade safety for sensitive business operations.
8Internal Efficiency Gains
Internal testing revealed significant efficiency improvements, with pull-request review costs dropping 32% as agents learned. By running 51 agents across 11 teams, Labelbox demonstrated that agent performance improves rapidly through memory updates. This internal rollout confirms the platform's ability to handle complex tasks like software engineering, finance, and procurement with measurable cost-effectiveness.
Original article: Unite.AI — Labelbox Secures $100M Series E to Scale AI Agents and RL