MICROSOFT COMMERCIAL MARKETPLACE·ICO-REGISTERED·AZURE UK SOUTH·DPIA-READY
MARKING · LIVE · MICROSOFT MARKETPLACE
Your marking, calibrated to your standard.
Consistent marking at scale, with a full audit trail for every score. Calibrated to your mark scheme, not a generic rubric. Built for UK universities, deployed on Azure UK South.
It is not a generic AI marker. It is your marker, calibrated to your judgement.
DELTA TO YOUR MARK SCHEME, BY BATCH
B1
B2
B3
B4
B5
B6
▼ CONVERGES WITHIN 3 BATCHES
WATCH IT EXPLAINED
See the Marking Agent in action.
WATCH ON YOUTUBE
How the Marking Agent calibrates to your mark scheme
A walkthrough of upload, grade, correct, and converge.
Free pilot access for institutions · Microsoft Commercial Marketplace · Azure UK South
USE CASES
Three ways institutions use it.
01
Bulk first-pass marking
Upload 200 submissions in one batch. The agent grades each against your mark scheme with per-criterion scores, an overall mark, and a written justification. You correct what needs correcting, and it learns from every correction.
02
Calibration across a team
When five markers grade the same paper differently, students notice. Run the agent across your team's marking history. It surfaces criterion-level inconsistency and gives the team a shared reference point.
03
Transparent second marker
Flagged disagreements, evidence per criterion, audit trail per paper. External examiners read the rationale directly, and disagreements surface for human review before any score is published.
HOW IT WORKS
Six steps to a calibrated marking partner.
01
Upload
Mark scheme, manual or PDF.
02
Submit
Bulk upload of papers.
03
Grade
Per criterion, every paper.
04
Correct
You adjust where needed.
05
Learn
Standards lock in.
06
Converge
Gap closes per batch.
MANUAL MARKING VS THE AGENT
Three concrete savings.
METRIC
MANUAL MARKING
WITH THE AGENT
Time per paper
~30 minutes
Under 2 minutes
Calibration to your standard
Weeks of training
3 batches of corrections
Audit trail compilation
Hours per module
Seconds per export
Consistency across markers
Varies marker to marker
Deterministic (temp = 0)
Directional figures from product capability and pilot data. Verified per-institution figures will follow as we close pilot data with our partners.
BUILT FOR TRUST
Research-grade tooling, UK infrastructure.
Data residency
Azure UK South. Data does not leave the UK.
Temperature = 0
Deterministic grading. Identical input gives identical output.
Audit trail per paper
Every score, rationale and correction logged for academic integrity.
No public retraining
Your data and your students' IP are never fed back to the foundation model.
COMMON QUESTIONS
What institutions ask first.
Will the agent grade exactly like I do?
Not on day one. It grades against your mark scheme out of the box, then calibrates as you correct its drafts. Most markers see meaningful convergence by their third batch.
Is the agent a black box?
No. Every score comes with criterion-level evidence, a written justification, and an audit trail of corrections. External examiners read the rationale without translation.
Can I trust it on borderline cases?
It surfaces low-confidence grades for human review before they ship. You see "Awaiting Calibration" or "Calibrating" status badges. Explicit, not hidden.
What happens to my data if I stop using it?
Pilot data is preserved in a read-only state when access ends. Your work is not deleted, and the sub-processor list is public.
ASSET PACK
Take it to your committee.
Ready to see it on your own marking?
We work with UK university partners on pilots and R&D collaborations. Contact us to scope what is right for your institution.
Start a free trial
hello@excoms.ai
ExComS.ai
Suite A, 1st Floor, 2a Windsor Street,
Luton, Bedfordshire, LU1 3UB, UK
Luton, Bedfordshire, LU1 3UB, UK
© 2026 ExComS Innovation Center Ltd. All rights reserved.
