Most formative assessment tools measure whether students can recall information under test conditions. This tells educators very little about whether a student has genuine conceptual understanding. A student can memorize definitions and pass a quiz without being able to explain why something works.
The Feynman Technique — the idea that you only truly understand something if you can teach it simply — is widely respected as a learning and self-assessment method, but has never been systematically applied as a structured, scalable assessment tool in real classrooms.
Axiom Flow was created to close that gap. The platform turns teaching into a measurable act. Students explain concepts to Sam, an AI student who only knows what it is taught. Sam is then examined on that taught knowledge. The exam result becomes the formal measure of how well the student understood and transferred the concept. This removes the guessing, surface-level recall, and test anxiety common in traditional formats — and replaces them with a direct measure of understanding quality.
An instructor uploads lecture notes, pastes academic content, or provides a YouTube video URL. Atlas, the assessment AI, analyzes this material and generates a set of misconceptions — structured incorrect beliefs about the topic — along with corresponding exam questions mapped to each one.
Sam, the AI student, is initialized with these misconceptions as his starting understanding. The student then teaches Sam the concept in a free-form conversation. Sam asks clarifying questions when explanations are unclear and updates his understanding solely based on what the student says. No external knowledge is used.
Once teaching is complete, Sam takes a structured mock exam. Each question maps to one misconception. Sam answers only from his taught understanding. Atlas then evaluates the answers, producing a score, identifying remaining misconceptions, and generating a breakdown of conceptual gaps.
The platform is deployed via Moodle LTI and is currently active with over 100 students in real coursework environments. Instructors receive structured assessment data without requiring manual grading of open-ended explanations.
Axiom Flow launched as a Moodle LTI integration and is currently in active use in real academic environments with over 100 students. It is production-ready and embedded directly into coursework as a formative assessment tool.
In the near term, the platform is expanding its LMS integrations to include Blackboard and Clever, broadening reach beyond Moodle-based institutions. The core architecture is designed to be LMS-agnostic, making integration into any major learning management system straightforward.
Over the next 2-3 years, Axiom Flow aims to establish itself as a standard formative assessment layer for higher education and secondary institutions — a scalable infrastructure for measuring conceptual mastery through teaching, not testing. The goal is to make Feynman Technique-based assessment as accessible and routine as quiz-based tools, while producing significantly richer data on student understanding.
Visit axiomflow.app to learn more. Axiom Flow is available as a Moodle LTI integration. Instructors can connect it directly to their Moodle course. For access, onboarding, or institutional inquiries, contact the team through the website.
