DeweyLearn has raised $5 million in an oversubscribed Series A funding round to expand its multimodal artificial intelligence platform for evaluating real-world skills and delivering expert-level feedback at scale. SJF Ventures led the round, with participation from Catalysis Capital, Morningside, Owl Ventures, and other investors.
The company plans to use the new funding to expand its technology across clinical and healthcare education, higher education, workforce training, and K-12 learning environments.
DeweyLearn’s platform combines audio, video, and educational data with domain-specific expertise and learning science. The technology is designed to observe how people perform tasks in physical or online settings and provide feedback traditionally delivered by instructors, evaluators, or industry experts.
The company is addressing growing demand for performance-based assessments that measure what learners can do rather than relying primarily on exams, written assignments, or self-reported knowledge.
These assessments are especially important in fields where competence depends on observable actions. A culinary student may need to demonstrate proper knife technique, while a nursing student must demonstrate the ability to perform clinical procedures accurately and safely.
Evaluating these skills typically requires an instructor or experienced practitioner to watch each learner individually. This can be difficult and expensive to scale across large classes, distributed training programs and institutions facing faculty shortages.
DeweyLearn uses multimodal AI to analyze the visual, audio and contextual information generated during a learning activity. Its system can evaluate the steps taken by a learner, compare them with the institution’s standards and generate actionable feedback.
The platform is trained using each customer’s curriculum, assessment criteria and subject-matter expertise. This enables DeweyLearn to adapt its analysis to different fields and institutional requirements rather than applying the same generic evaluation model to every learning environment.
The company believes this approach can expand access to high-quality feedback while allowing instructors to focus more of their time on personal guidance, complex interventions and student engagement.
DeweyLearn recently won the 2026 ASU+GSV Cup after being selected from more than 3,000 companies as the competition’s leading education technology startup.
One of the company’s primary deployments is at the Auguste Escoffier School of Culinary Arts. DeweyLearn has evaluated more than 20,000 student homework submissions for the institution.
The platform serves as an initial reviewer, with instructors approving or modifying the resulting assessments. DeweyLearn said the system has saved hundreds of hours that faculty members would otherwise have spent grading assignments.
Students can also receive feedback more quickly instead of waiting for an instructor to manually review every submission. The time saved allows faculty members to devote additional attention to individual students and more advanced instructional needs.
The company is also working with Riverside Insights, a provider of specialized student assessments. Riverside is exploring how DeweyLearn’s technology could improve the efficiency and effectiveness of its assessment products.
DeweyLearn’s applications extend beyond traditional academic settings. The platform is being used in clinical training by the NeuroAffective Relational Model, which provides continuing education for therapists treating developmental and complex trauma.
Clinical training is a particularly relevant use case because instructors must evaluate not only whether a learner understands a concept but also how effectively that person applies it during realistic interactions.
The platform could help educational organizations review simulated patient encounters, therapeutic sessions, technical procedures and other complex activities that involve speech, movement, decision-making and interpersonal behavior.
DeweyLearn believes multimodal AI can support any learning environment where an expert would normally need to see, hear and interpret a learner’s performance.
Potential applications span healthcare, hospitality, information technology, skilled trades, professional development and other industries in which workers must demonstrate practical competence.
The technology may also help employers and educational institutions develop more consistent assessment standards. Human reviewers can vary in how they interpret performance, particularly when evaluating large numbers of learners over extended periods.
An AI system trained on clearly defined standards can provide an initial assessment using the same criteria across submissions. Human instructors can then review the results, make adjustments, and focus attention on situations requiring more nuanced judgment.
DeweyLearn was founded by CEO Luyen Chou and CTO Dirk Liebich.
Chou has held several leadership positions in education technology. He previously served as chief learning officer at 2U, chief product officer at Pearson, and founder of The School at Columbia University.
Liebich brings experience in applied AI, data analytics, and predictive modeling. His work at DeweyLearn focuses on developing systems that can understand learning activities across different subjects and environments.
The company is named after philosopher and education reformer John Dewey. During a trip to China between 1919 and 1921, Dewey encouraged Chou’s grandmother to study to become a teacher at the University of Chicago.
DeweyLearn creates a tailored knowledge graph for each customer by mapping the skills, behaviors, and performance indicators relevant to a specific learning domain.
For example, a culinary knowledge graph could define the movements and outcomes associated with a particular knife cut. A clinical training model could identify the actions, communication patterns, and procedural steps required for a successful patient interaction.
As the system observes additional learning activities, it can expand its understanding of how different actions relate to performance and educational outcomes.
DeweyLearn is also developing a broader model of human learning based on repeated cycles of observation, intervention, and measurement across multiple fields.
The company believes this accumulated knowledge could eventually help institutions understand not only whether a learner completed a task correctly but also which instructional approaches are most effective for improving performance.
The Series A financing will support continued product development and broader adoption among schools, healthcare training programs, employers, and other organizations seeking scalable skills assessments.
By combining multimodal AI with institution-specific expertise, DeweyLearn aims to make detailed performance feedback available to more learners without attempting to remove human instructors from the educational process.
KEY QUOTES:
“Human expertise has been a limited resource for our entire history. With multimodal AI, DeweyLearn can give an aspiring chef real-time feedback on her knife technique from the world’s greatest chefs. It can assess the effectiveness of clinical therapists, as well as allow nursing students to receive real-time feedback in simulated hospitals. We’re making the kind of expert feedback that once required a master watching over your shoulder available to every learner. DeweyLearn represents a paradigm shift in how we assess skills and apply human expertise at scale in areas as varied as healthcare, hospitality and IT.”
Luyen Chou, Co-Founder and CEO of DeweyLearn
“Riverside, a provider of specialty assessments to students, is working with DeweyLearn to explore new approaches to its assessment offerings to improve efficiency and effectiveness.”
Katy Genseke, Psy.D., Head of Clinical Product at Riverside Insights
“SJF is focused on enabling better learning outcomes and career opportunities for all. We were wowed by the transformative potential of DeweyLearn’s approach to applying multimodal AI for education and confident that the founders have both the industry experience and AI expertise to execute on that potential. We’re proud to partner with DeweyLearn to help instructors and learners master the science of education.”
Arrun Kapoor, Managing Director at SJF Ventures
“DeweyLearn is building a model of human learning informed by real cycles of observing, intervening and measuring across diverse learning domains. Our novel approach quickly builds a knowledge graph tailored to each customer, like a knife-cut technique or what makes a student chef successful. But in doing so, we’re also creating a meta knowledge graph of how people learn. Much like Google Earth, we’re building a world model of learning that empirically understands learning at an action level and can apply the insights at scale.”
Dirk Liebich, Co-Founder and CTO of DeweyLearn
