Education Technology

Learning platforms that can tell who is actually learning

We make progress and assessment models explicit, then build the feedback loops that show instructors where a learner needs help while there is still time to give it.

Where engagements usually start

  • Course and cohort delivery
  • Assessment and credentials
  • Adaptive pathways
Illustration of an open book, a graduation cap, a progress ring, and a lesson player card

Mastery-based

progress measured against evidence

WCAG 2.2 AA

accessible across devices and abilities

Early signal

at-risk learners visible in week two

The hard part

Recording activity is easy, proving learning is not

A learning platform serves students, instructors, administrators, and institutions, and each measures success differently. Most systems record activity because activity is easy to capture, then report completion as if it were understanding. We define the mastery model first, instrument against it, and design intervention tools so the insight reaches the person who can act on it.

  • Completion is not comprehension

    Video watched and modules finished are easy to record and tell you very little. Without an evidence model, reporting flatters everyone and helps nobody.

  • Four audiences, one product

    Learners want momentum, instructors want signal, administrators want compliance, institutions want outcomes. A single undifferentiated interface serves none of them well.

  • Access is uneven

    Learners arrive on shared devices, slow networks, and assistive technology. Anything that assumes a modern laptop excludes part of the cohort.

What we build

The units of a learning platform

Each one can be built on its own or added to what you already run. Together they cover delivery, evidence, and the institutional plumbing around both.

Unit 01

Course and cohort delivery

Curriculum structure, scheduling, live and self-paced modes, and the enrolment logic institutions actually need.

  • Curriculum
  • Cohorts
  • Enrolment
Unit 02

Assessment and credentials

Item banks, rubrics, integrity controls, and verifiable credentials tied to demonstrated evidence rather than attendance.

  • Item banks
  • Rubrics
  • Verifiable credentials
Unit 03

Adaptive pathways

Learner models that adjust sequence and difficulty from real performance, with the reasoning visible to instructors.

  • Learner model
  • Sequencing
  • Mastery
Unit 04

Instructor insight and intervention

Dashboards built around the next action, surfacing which learners need contact this week and why.

  • Risk signals
  • Cohort view
  • Nudges
Unit 05

Institutional integration

LTI, SIS, roster sync, and single sign-on so the platform fits the systems a school or employer already runs.

  • LTI 1.3
  • OneRoster
  • SSO
Unit 06

Accessibility and inclusion

Screen-reader support, captioning, keyboard paths, and low-bandwidth modes designed in rather than remediated.

  • WCAG 2.2 AA
  • Captions
  • Low bandwidth
Technology

What we build learning platforms with

Chosen for the device a learner actually has, for assessment data that stands up to analysis, and for the standards institutions expect you to speak.

Learning surface

Works on the device a learner has

  • Next.js
  • React
  • TypeScript
  • React Native
  • Radix UI

Platform services

Structure, assessment, and identity

  • Node.js
  • Python
  • PostgreSQL
  • Redis
  • Keycloak

Content and media

Delivered cheaply at scale

  • Mux
  • Cloudflare Stream
  • S3
  • SCORM / xAPI

Analytics and modelling

Turning activity into evidence

  • dbt
  • BigQuery
  • Airflow
  • scikit-learn
  • Caliper
How we work

The order we run an engagement in

Every stage produces something the next one needs, which is why the mastery definition comes before a line of tracking code.

  • Stage 1 of 4

    Define what mastery means

    We agree the evidence that counts as learning for your subject before deciding what the product will record.

  • Stage 2 of 4

    Instrument the learning event

    Interactions are captured against a standard model, so analysis is possible without re-instrumenting later.

  • Stage 3 of 4

    Close the loop to instructors

    Signals are turned into a weekly action list, and we measure whether intervention actually changed the outcome.

  • Stage 4 of 4

    Prove accessibility and scale

    Assistive-technology testing and load rehearsal for enrolment peaks happen before launch, not after the first term.

What changes

What a cohort and its instructors get back

  • Progress measured against mastery, not minutes
  • At-risk learners visible while there is time to help
  • Accessible journeys across devices and abilities
  • Institutional reporting produced without manual work

Standards we build to

  • WCAG 2.2 AA
  • LTI 1.3
  • OneRoster
  • xAPI
  • FERPA
  • GDPR

Systems we integrate

  • Canvas
  • Moodle
  • Blackboard
  • Google Classroom
  • Clever
  • Zoom
  • Stripe
  • Credly
Questions

What education teams ask us first

  • Usually not the whole thing. Enrolment, delivery, and grading are commodity. The defensible part is your pedagogy, your assessment model, and your data. We often build that as a product that integrates over LTI with whatever LMS an institution already runs.

Build a platform that can prove it works

Tell us what mastery means in your subject and where the current product stops short. We will come back with an evidence model and a first release plan.