Encord Raises $60M for AI Data Infrastructure
AI data infrastructure company Encord just secured a $60 million Series C round led by Wellington Management, bringing its total funding to $110 million. The company, which builds tools for managing the complex data needed for 'physical AI' models, reported 10x revenue growth over the last year. The funding is aimed at scaling its platform as AI moves from digital spaces into physical applications like robotics and science.
Encord was founded by co-CEOs Ulrik Stig Hansen and Eric Landau after they observed machine learning teams spending upwards of 80% of their time on data preparation rather than model development. This latest funding round brings the London-based company's post-money valuation to $550 million. The company's focus on "physical AI" addresses systems that interact with the real world, moving beyond digital-only applications. For biomanufacturing, this translates to powering the next generation of lab automation, where robots and sensors must interpret complex, unstructured data from their immediate environment—a significant step beyond the capabilities of traditional LIMS. This funding addresses a core challenge in advanced biologics manufacturing: data standardization and integration. Platforms like Encord's provide the underlying infrastructure for creating digital twins of bioprocesses, allowing for *in silico* simulation and optimization of a viral vector production run before consuming expensive reagents and cleanroom time. Lead investor Wellington Management has a track record of backing disruptive technologies and has specific venture funds for both biotechnology and tech-enabled solutions like AI and data analytics. Their investment signals a recognition that robust, scalable data infrastructure is becoming a critical component for GMP-compliant operations and realizing the vision of Industry 4.0 in biopharma. Encord's platform is built to manage the entire data lifecycle for petabyte-scale, multimodal data—including video, sensor feeds, and specialized formats like medical DICOM imagery. This capability is crucial for cell and gene therapy, where process development, manufacturing operations, and analytical teams generate vast, disparate datasets that must be unified for regulatory compliance and process improvement. While competing with data-labeling services like Scale AI and Labelbox, Encord differentiates itself by focusing on providing a unified data layer that combines curation, annotation, and model evaluation. This integrated approach is designed to create a continuous feedback loop, where models help improve the quality of the data that trains them—a key factor for adapting automation in dynamic manufacturing environments.
