LXT Launches Crowd-as-a-Service API for Scalable AI Data Workflows
Ten million contributors are now accessible directly through enterprise systems, as Toronto-based LXT unveils its Crowd-as-a-Service model. By bypassing traditional outsourcing marketplaces, the platform allows AI teams to integrate global human input directly into their own technical environments, maintaining end-to-end control over task design, quality standards, and project delivery timelines.

The offering, which leverages the infrastructure gained through LXT's acquisition of clickworker, functions as a direct API connection to a massive, vetted contributor network. Unlike legacy platforms that operate as passive marketplaces, LXT manages the technical scaling and contributor oversight, while the client retains authority over internal workflows. This ISO 27001-certified, GDPR-compliant architecture addresses common enterprise bottlenecks in data collection, annotation, and reinforcement learning with human feedback (RLHF).
For teams developing large language models or computer vision systems, the service provides access to participants across 150 countries and over 1,000 language locales. Mark Sewell, VP of Growth at LXT, noted that the model is designed to solve the persistent conflict between the need for high-volume human input and the desire for internal control. By embedding the contributor network into existing systems, organizations can accelerate tasks like speech recognition and search relevance rating without sacrificing the demographic or linguistic diversity required for robust AI performance.
Comments (0)
No comments yet. Be the first!