SAN FRANCISCO — Encord, an AI data development platform for advanced vision and multimodal teams, has announced a $30 million Series B funding. The round was led by Next47, a global venture firm investing in early and expansion-stage companies in SaaS, AI and enterprise, with participation from existing investors Y Combinator, CRV, and Crane Venture Partners.
The effectiveness of an AI model is tied directly to the quality of the data it is built on. As executives face increasing pressure to integrate AI into their enterprises, they often encounter the challenge of siloed, uncurated and unprepared data, which significantly hampers the model development process. This issue is a major bottleneck preventing broader AI adoption. Moreover, the expansion of AI highlights another critical problem: the scarcity of well curated high-quality data and the insufficient number of human domain experts needed to prepare this data and evaluate models for effective AI training and fine-tuning.
As opposed to solely relying on human supervision, Encord leverages automation to give AI teams everything they need to make large datasets AI-ready and get models to production faster. Encord’s software addresses all four steps of turning data into AI: data management, curation, annotation and model evaluation. By consolidating this entire workflow into one platform, companies have a trail that shows why a model is making certain decisions. This allows enterprises adopting AI to uncover weaknesses and easily determine the correct data that is needed to retrain the model for a better outcome. With a software-first approach, Encord easily scales with customers as their data volume and model output increases, addressing the limitations of other tools and purely human supervision.
“In 10 years, every company will have an AI department in the same way that they have an IT department today or risk losing relevance, and Encord will be the only platform they need to accomplish the process of turning data into functional AI models,” said Ulrik Stig Hansen, Co-Founder and President of Encord. “As AI models begin tackling increasingly complex problems, the access to and management of data will only become more crucial. By leading with automation, we’ve built a scalable solution to support customers on every step of their AI journey.”
To bring even more ease to the management and curation steps of the process, Encord is also launching Encord Index, an end-to-end data management tool that allows customers to visualize, search, sort, and control their internal data that will be used to train and create AI models. Index integrates seamlessly with data storage such as AWS S3, GCP Cloud Storage, Azure Blob and others to automate the process of curating the best data and removing uninformative or biased data. As a result, Encord customers have achieved a 35% reduction in dataset size by curating the best data, see upwards of 20% improvement in model performance and save hundreds of thousands of dollars in compute and human annotation costs.
“Successful state of the art models, like our recently released Expressive Avatar foundation model EXPRESS-1, require highly sophisticated infrastructure. Encord Index is a high-performance system for our AI data, enabling us to sort and search at any level of complexity. This supports our continuous efforts to push the boundaries of AI avatar technology and meet customer needs” said Victor Riparbelli, Co-Founder and CEO of Synthesia, the billion-dollar generative AI company.
The Encord platform was built on the foundation of vision, an inherently more complex modality than text, positioning the company to lead the imminent multimodal era, outpacing those who are currently solely focused on LLMs with the current hype cycle. Encord can manage data for AI models based in images, videos, medical imagery, visual language, voice and more. With more than 2,000 predictive and generative models put into production, Encord is trusted by over 120 of the top AI teams from organizations such as Synthesia, Philips, Zoopla, Cedars-Sinai, Northwell Health, and many more.
