Colleague AI Awarded $1.25-2M NSF SBIR Phase II Grant to Deepen AI Tools and Innovation in K-12 Education

Prestigious Award to Boost R&D Efforts Including Agentic AI Tools and Architecture Custom Built for K-12 With Extensive Co-design and In-School Studies

SEATTLE, WA, UNITED STATES, August 19, 2026 /EINPresswire.com/ — Colleague AI, a research-backed K-12 AI platform and spin-off from the University of Washington College of Education, today announced it has been awarded a Phase II Small Business Innovation Research (SBIR) grant from the National Science Foundation. The award provides $1.25 million in base funding, with supplemental funding expected to bring the total to approximately $2 million over the next 18 to 24 months.

The Phase II award builds on Colleague AI’s SBIR/STTR Phase I project which funded the development of four specialized AI models — for lesson quality measurement, retrieval and augmentation, instructional nudging, and subject-specific content generation — that now power the company’s lesson planning and grading tools for K-12 teachers.

The funding boost will cover the research, development, testing, and release of major tooling additions to the platform such as finely-tuned AI assessment and student feedback systems, an on-demand classroom-data AI analyst expert, three-actor learning tools for teacher-student-AI experiences, and advanced student AI tutoring with deep context and memory on subjects, curriculum, student competencies, and specific classroom assignments.

“This funding lets us lead the market in quality, which is what schools and districts want,” said Dr. Min Sun, co-founder and CEO of Colleague AI. “Our Phase I grant proved that AI, built with effective pedagogical grounding, can outperform other tools in the market both in quality and through research on AI’s impact on education. Phase II is about expanding today’s agentic abilities and our disciplined research-based approach as well as selecting districts who have specific requests for AI features that they want developed.”

Under the grant, the company will use collaborative research, co-design with K-12 stakeholders, extensive evaluation, thorough testing with school district alphas and betas, and better interoperability to develop additional features in grading, tutoring, student discussions, administrative tools, actionable data insights, and more. Colleague AI is actively seeking district partners with requests for new AI features and tools willing to co-develop under the grant as well, with options for additional funds up to $250K.

“Phase II development will also solidify us as a technical leader in this space by developing extensive data architecture that meets agentic capabilities, custom-built and tuned specifically for K-12 and focused on applying academic research. There are so many nuances in education that are absolutely vital to the quality, performance, and output of AI systems, and this is what we’re capturing and scaling.” said Jian (Kevin) He, co-founder and CTO.

The company has signed MOUs with 10 school districts to study the platform’s effects on instructional practice and student learning outcomes during the 2026-27 school year. Districts interested in participating in such research and grants should contact the company directly. Colleague AI will continue its academic research partnerships that help enforce rigorous checks, benchmarks, qualitative and quantitative analysis, and instructional best practices that result in features and tools that are higher quality than competitors.

The company has already begun new feature development and will extend access to platform additions across its partner districts and Pro subscribers as they become ready. Product announcements are expected later this year.

Colleague AI now serves over a quarter of million users nationwide and is available to schools, districts, and other learning organizations who serve K-12 students. Inquire for organizational partnerships or sign up for free trial at colleague.ai.

About Colleague AI
Colleague AI is the leading comprehensive K-12 AI infrastructure platform serving teachers, students, school administrators, and district leaders. The platform services the entire teaching and learning cycle, offering unlimited educator and administrative tools, easy-to-use embedded workflows, and AI management infrastructure. Built with more than $18 million in federal research funding from the NSF, U.S. Department of Education and Institute of Education Sciences, and developed in partnership with AmplifyLearn.AI at the University of Washington College of Education, Colleague AI serves 250K+ users nationwide and in 50+ countries. Learn more at Colleague.ai.

Min Sun
Colleague AI
email us here

Media gallery