Timing measured on one trainer’s batch of 24 assessments. Real evidence, not yet a cohort study. How we measured it.
An assistant to the judgment, not a machine that marks.
Most platforms mark only what’s typed into their question fields. VET assessment lives in templates, portfolios and evidence documents. Apollo reads those, prepares the judgment, and leaves the decision where the Standards put it.
What the platforms still can’t do.
Feedback that quotes the work
Comments name the guide’s own requirement, point at what the student wrote, and say exactly what to add.
It knows which kind of task it is
Apollo tells a scenario task from a knowledge paper by reading the task’s own paperwork, and marks each the way it should be marked.
It reads real documents
Apollo reads the file the student actually submitted, and returns it marked up for the assessor.
Nothing is ever stored
Apollo has no server and no submissions database. Student work is cleared the moment the assessor closes it.
Integrity checked on your machine
Every batch is compared for copying, against classmates and the sample answer, without uploading anything anywhere.
Your AI account, at cost
No per-learner pricing wrapped around it, and no margin added by us.