A Saudi assessment platform

AI-driven online assessment for educational institutions

From the item bank to a verifiable certificate — Arabic-first, hosted in-Kingdom, with reports ready for the accreditation file.

Arabic-firstHosted in-KingdomAccreditation-alignedContinuous psychometricsAI with stated limits
01

Test thousands of students at once without anything breaking

Finals day tests the platform, not the student. Every interaction is persisted instantly; if the network drops, answers continue locally and reconcile when it returns; and a session resumes on another device from the exact second it stopped.

Offline-first · session recovery · single-device enforcement

Session status
1,847active sessions
Response time82ms
Offline sessions12
All answers persisted
02

Know which of your items are broken before they cost a student

An item that strong students get wrong more often than weak ones is not hard — it is broken. The platform computes facility and discrimination for every item from real responses and surfaces the unwell ones before they affect a decision.

Automatic item analysis · distractor analysis · atomic cohort rescoring

Item 7 quality
Facility0.42
Discrimination−0.13
Option C2%
Suspend and review the key
03

Produce accreditation evidence from daily operation, not at year end

Every item is linked to a learning outcome at authoring time, inheriting its Bloom's level and domain automatically. The attainment report exports in one click, documented with its threshold, method and framework version — not assembled by hand weeks before a visit.

PLO → CLO → item → response · export ready for the accreditation file

Accreditation export
Outcomes4 / 4
Mapped items128
Framework version2020
Ready to export

Artificial intelligence at six practical points

Not a banner on a homepage but six capabilities working inside the assessment cycle — each with a stated boundary that keeps the final decision human.

Item generation from your content

Upload a course file or paste text, set difficulty and outcome — items are generated already linked to that outcome and land in drafts for review.

Rubric building

Proposed criteria and performance levels for essay items, edited by the instructor before approval.

Essay scoring

Evaluation against the approved rubric with a proposed score and per-criterion reasoning — approval stays human.

Proctoring and identity

Continuous facial matching, multiple faces and gaze deviation, and detection of identical answer patterns.

Quality reports

Strengths and areas for improvement per course, articulated from attainment and item analysis data.

Adaptive testing and early warning

Higher precision with fewer items, and flagging a student at risk while intervention is possible.

AI proposes, analyses and prioritises review. Integrity verdicts and final grades remain human decisions attributed to a person in the audit trail. Read the boundaries in full

Why it is built here rather than imported

Global platforms support Arabic by translating the interface, leave data tables left-to-right and certificates with disconnected glyphs, and host your data abroad.

Accreditation-aligned

Outcome structures and learning domains follow the adopted course-specification templates, and are configurable when the framework changes.

Inside the Kingdom

Databases and proctoring media are hosted locally by default, not as a paid upgrade.

Arabic-first

Full layout mirroring, Arabic fonts embedded in certificate and report generation, and an editor that handles Arabic prose alongside code.

Published methodology

Eight reference articles and a 154-term glossary — published to be scrutinised, not believed.

How you start

With one scope that works well, not an institution-wide project that takes a year.

Step 1

A demo on your data

We take one real course or programme of yours and walk it in front of you — not a canned deck.

Step 2

A pilot in one scope

One course or one training track, with an approved outcome-mapped bank, in weeks rather than terms.

Step 3

Scale on real data

After the first assessment you hold real numbers on your own item quality — the best basis for a decision to expand.

Ask for a demo on one of your own courses

We take a real course or training track and walk it from the item bank to the attainment report and item analysis — live, in front of you.