Skip to content
Industries

Learning that works without a connection.

iLeaf has built education technology since 2011 — LMS and LXP platforms, interactive e-books, and offline-first mobile learning that syncs when a connection returns. We added machine learning to reduce dropout in US schools, and Scandlearn's compliance training runs on our engineering.

Ask Lia about education

What this includes

LMS & LXP platforms
Course delivery, cohorts, assessment and reporting, including compliance-driven training environments.
Offline-first mobile
Learners keep working without connectivity, with conflict-aware sync when they reconnect.
Adaptive learning
Paths that respond to how a learner is actually doing, rather than a fixed sequence for everyone.
Dropout prediction
Early-warning models on engagement signals, so intervention happens while it still helps.
Interactive content
E-books and rich media that bridge printed material and digital delivery.
AI tutoring & marking
Assistance grounded in your own curriculum, with teachers reviewing rather than being replaced.

How we run it

Every phase ships something usable on its own, so you are never holding a half-finished system waiting on the next milestone.

  1. Design for the worst network

    Offline is the default assumption, not a later feature. Retrofitting sync into an online-only product is painful.

  2. Instrument learning, not clicks

    The signals worth capturing are progress and struggle, which is what early-warning models depend on.

  3. Keep teachers in the loop

    AI drafts and flags; educators decide. Anything else erodes trust with the people who have to use it daily.

  4. Ground it in the curriculum

    Retrieval runs over your own material, so answers match what is actually being taught and assessed.

What we build it with

  • TypeScript
  • React
  • Node.js
  • Swift
  • Kotlin
  • Flutter
  • SQLite
  • PostgreSQL
  • Python
  • SCORM / xAPI
  • AWS
  • Graph-RAG

Questions we get asked

How does offline learning actually sync?

Content is cached on the device and progress is written locally, then reconciled when connectivity returns. The hard part is conflict handling — two devices, one learner, divergent state — which is why we design the data model for it from the start rather than adding sync later.

Is AI tutoring reliable enough for students?

When it is grounded in your curriculum rather than answering from general knowledge, yes — and it should cite the material it drew on. Teachers review flagged work rather than being replaced, which is both better pedagogy and considerably easier to get adopted.

What education experience do you have?

Since 2011: LMS and LXP platforms, Scandlearn's compliance training, interactive e-book applications, offline-first mobile learning, and machine learning applied to reducing dropout rates in US schools.

Let’s talk about education.

Tell us what you are running and what it needs to do next. We will tell you honestly whether we are the right team for it.

Talk to a solutions lead
Selected work

Education: the work behind this