Your coursework goes in. Assessments your faculty approved come out.
Six stages, in order. Nothing reaches a learner that a person at your institution has not signed off on — and the same corpus opens a self-paced practice track for your students.
- 01
Ingest your coursework
Chapters, syllabi, prior assignments, competency frameworks. Everything downstream is generated from this corpus, so the assessments speak in your institution's language rather than a generic bank's.
- 02
Draft the items
The engine writes questions across types and difficulty bands, each one tied to a learning objective. It drafts the rubric alongside them, because a question without a scoring standard is not an assessment.
- 03
Your people approve
Teachers and faculty accept or reject each item. Nothing reaches a learner unapproved. The surviving set becomes the institution's question bank and it accumulates term over term.
- 04
Deliver adaptively
Each answer picks the next item. Strong learners get harder questions instead of more of them; struggling learners get the sub-skill they actually missed.
- 05
Report what happened
Results by learner, cohort and objective, in time to act on them. Administrators see usage, volume and coverage across the institution.
- 06
Inside your walls
Client-isolated deployment, encrypted in transit and at rest, audit trails throughout. Built to meet FERPA, GDPR and SOC 2 obligations.
The same six stages serve two people.
A teacher walks the pipeline forwards — ingest, draft, approve, deliver, report. A student walks in at the middle of it, practising against the coursework the teacher approved from, on their own schedule and with nothing riding on it.
Shared foundation
Everything starts from your own material.
Chapters, syllabi, prior assignments, competency frameworks — ingested once. Both sides of the platform are generated from that same source, so what a teacher sets and what a student practises are the same subject, in the same language, held to the same standard.
For educators
Set the standard, then hold the work to it.
- Draft items straight from a chapter you already teach, across question types and difficulty bands
- Approve or reject each one — nothing reaches a learner unapproved
- The rubric is written alongside the questions, not invented afterwards
- Results by learner, cohort and objective, back in time to change the lesson
For students
Practise the same material, at your own pace.
- Self-paced sets drawn from the courses you are actually enrolled in
- Multiple choice, fill in the blank, short and long answer — every one with its worked answer
- Difficulty follows you: more of what you missed, less of what you have already shown
- No grade, no deadline, no record. Practice belongs to the student
Held on both sides
Practice and assessment are drawn from one generation ledger and never overlap. A learner can work through everything the platform will ever give them and still not have seen the exam — which is what makes it safe to open both doors onto the same coursework.
What you get in each of those stages.
Automated creation
Point the engine at a chapter, a syllabus or a competency framework and it drafts the items. Your faculty approve or reject each one; what survives becomes the bank.
Personalised at scale
Difficulty, phrasing and sequence shift with each learner's history. A cohort of ten thousand sits ten thousand tests, marked on one scale.
Complexity diagnosis
When a learner stalls, the engine narrows down which sub-skill is missing rather than reporting a low score and leaving you to guess.
Every question type
Multiple choice, fill in the blank, short answer and long answer — on laptop, tablet or phone.
Bias review
Items are checked for linguistic, cultural and contextual load before they reach a learner, and flagged for human review when the check is uncertain.
Real-time insight
Results land as they happen, by learner, by cohort and by objective — so a gap in week two is visible in week two.
Fits your stack
Runs on its own, or integrates with any existing LMS as the institution needs. API-first and multilingual by default.
Improves with use
Items that fail to separate strong learners from weak ones are retired automatically. The bank sharpens every term.
A complete system on its own. Connects to yours if you need it to.
AptitudAI runs standalone, with its own teacher, student and office views, so an institution needs nothing else to assess. Where a school already lives in an LMS, it integrates with that system as the institution needs: grades, rosters and results flow both ways through the API.
- Canvas
- Moodle
- Blackboard
- Microsoft Teams
- Any LMS with an API
Deployed inside your environment, isolated per client.
Encrypted in transit and at rest, with audit trails on every action. We set the deployment up once; after that your team runs it. Student data is handled as set out in our Privacy Policy: the institution is the controller, and we never sell it or use it to serve anyone else.
Working toward compliance with
- FERPA
- COPPA
- GDPR
- SOC 2