SceneSnap transforms existing training materials—slides, PDFs, and lecture recordings—into adaptive, interactive learning experiences through a multi-agent AI system called Repeater. The platform measures actual learning outcomes rather than just completion rates, providing management-ready evidence of skill acquisition. It integrates with major LMS platforms via LTI 1.3, SCORM, and SCIM 2.0, serving over 40 universities and companies with 1.6 million active learners.
Funding
Funding not disclosed
Founders
Product
Problem
Training delivered is not training learned. Missed live sessions stay missed, leaving skills gaps and collapsed completion rates. Completion metrics fail to answer whether learners actually understood the material, making L&D budgets vulnerable to cuts. Interactive content traditionally takes weeks to build, requiring resources most teams lack.
Solution
SceneSnap provides a multi-agent AI learning system called Repeater that converts existing materials—recordings, PDFs, slides, and transcripts—into adaptive, interactive learning paths. The system uses a pipeline of four specialized agents: a Planner that decomposes source material into an ordered learning plan, an Explainer that teaches each concept in the optimal modality, an Evaluator that probes understanding with quizzes and open questions, and an Adaptive agent that identifies gaps and builds personalized follow-up sessions. Every interaction generates evidence of learning, which is compiled into management-ready dashboards. The platform integrates with existing LMS and identity infrastructure, allowing learners to access experiences without a second login.
Target Audience
Primary customers are universities, professional certification programs, and corporate L&D teams that need to convert existing training materials into measurable, adaptive learning experiences and demonstrate learning outcomes to management.
Features
- Multi-agent pipeline architecture with typed contracts between Planner, Explainer, Evaluator, and Adaptive agents, ensuring inspectable and recoverable decision-making
- Automatic conversion of PDFs, slide decks, lecture videos, and transcripts into structured learning plans with editable modules and steps
- Multimodal content generation including narrated videos, mind maps, flashcards, animations, and comparison cards selected per concept
- Evaluator that compares open-ended answers against source material, returns structured feedback, and re-explains missed parts without gating progression
- Adaptive session generation that reads interaction signals—wrong answers, skipped steps, hesitations, time-on-step—to target weak spots in subsequent sessions
- LTI 1.3/LTI Advantage and LTI 1.1 support for single sign-on, deep linking, roster sync, and gradebook results
- SCORM package export for LMS-hosted content and SCIM 2.0 for automatic user and group synchronization from identity providers
- Enterprise SSO via SAML or OIDC with restricted origins and instant deactivation controls