The learning management system for universities — and, from the same team, POTECG for ECG interpretation training.
Built around how students learn.
Courses as pathways of lessons and practice, written with an AI assistant that only proposes, and practised with spaced repetition, honest progress and fifteen exercise types. Each university gets its own branded site with its own subject modules.
Built for universities: faculties, departments and the teachers who run the courses.
Request an AdaptIQ demoA simulated walk through AdaptIQ. A new quiz appears on a student’s pathway; the student answers a question and marks the QRS complex on an ECG.
Upload what you already teach from. The assistant reads it, proposes a pathway and writes lessons and exercises step by step. Nothing reaches a student until a teacher has accepted and approved it.
A simulated walk through AdaptIQ. A teacher attaches a lecture in the AI chat and asks for a four-question quiz. The AI drafts it, including a question on an ECG with its answer key; the teacher moves one wave label the AI placed wrong and accepts the exercises.
PDF, Word, PowerPoint, Excel, CSV, Markdown or images. Materials are extracted, indexed and searchable; a vision model describes the images.
The AI reads a document’s outline and pages and proposes topics, lessons and practice sets — each with the page range it comes from.
“Generate content” writes every lesson in the material’s language and 3–8 exercises per practice set. A progress bar shows how far it has got and what it has cost so far. You can stop it at any time; a step already running still finishes.
Accept one proposal, everything below it, or a whole batch. Ask for a change and compare before and after. Try any exercise in the real student player first.
Accepted AI content is marked as AI-made and waits for a teacher’s approval. Publishing is refused while unapproved AI content remains.
Twelve built-in types, two from the 3D atlas module and one from the biomechanics module. The AI assistant can draft the eleven marked with a spark. AI can draft
Subject modules plug into AdaptIQ with their own exercise types, tools and pages. Here is what the ones we built can do.
A WebGL viewer with 2,234 meshes and 3,432 named structures, searchable by name. It powers two exercise types and opens from inside lessons and practice.
Built from BodyParts3D, © The Database Center for Life Science (DBCLS), licensed under CC BY 4.0.
Headings, tables, resizable images with captions, quotes — and formulas typeset with KaTeX, inline or as blocks, inserted through a formula dialog.
We turn a content package — Markdown lessons plus exercises — into a module with one command, and updating it keeps student progress. Deeper modules add their own exercise types and tools.
Each problem is generated from a seed, so every student and every attempt gets its own numbers. The server rebuilds the problem and recomputes the grade; answers count within ±1.5%.
Find the vertical support reaction at A.
Find the vertical support reaction at A.
AdaptIQ can run as its own branded site for your university, with the controls a university needs around content, people and AI.
Each site has its own name, tagline, favicon and subject modules, and its own database and object store.
SAML 2.0 single sign-on support for university identity providers. New SSO accounts wait for an administrator’s approval.
A module can name the teachers who may edit it; other teachers cannot change it.
Every teacher or AI save of a lesson or exercise is stored as a revision that an administrator can restore.
A daily AI budget per site that fails closed, per-teacher rate limits, and the cost of every content build shown as it runs.
Uploaded text is treated as data, never as instructions. AI output passes an allow-list filter, and AI actions are audit-logged.
The assistant’s 22 tools form a Model Context Protocol registry, served as JSON-RPC 2.0 and bound to one conversation and the teacher’s module rights.
Each site runs as its own self-hosted stack — PostgreSQL, Redis, Qdrant and an S3-compatible object store — with a separate database and object store for every site.
Module accent colours from eight presets, each checked for 4.5:1 contrast. Light, dark or system appearance, set per user.
ECG interpretation training for medical education.
Book a demo and we’ll walk you through the AI workspace, the student player and the review flow — on a phone and on a desktop.
Request a demo