It works beside the systems you already run.
Keep your current systems. Apollo just reads the files you already have.
Apollo against a typical RTO platform.
| Apollo | A typical RTO platform | |
|---|---|---|
| What its AI reads | Your files as handed in: Word, Excel, PDF, scans | Answers typed into its own forms |
| Who makes the judgement | Your assessor | Often a suggested result, accepted in bulk |
| What the feedback says | Quotes the work and says what to fix | Often just a result, with the wording left to you |
| Where student work sits | Not kept by Apollo | On their servers |
| Who pays for the AI | Included in your fee | Built into their subscription |
| Integrity checking | Compares the class, on your computer | Generally not offered |
| To adopt it | Sign in from a browser. Nothing to move over | Data migration and re-authoring assessments |
| What a trainer learns | One screen | A whole platform |
| To stop using it | Cancel any time. Nothing to take back | Your records live in their system |
Why we call it a VET Education Co-Pilot.
Built for VET, works for the trainer, and helps rather than decides.
VET. It reads the real evidence
Apollo reads what the learner handed in, against your marking guide.
Trainer. The decision stays with them
Under the Standards, the assessor decides competence. Apollo can’t.
Co-pilot. It does the hours in front of the decision
It does the reading and drafting, so you have more time to judge.
And the record says so
Each returned document shows who marked it, and when.
What Apollo is not.
It’s a marking instrument, and deliberately nothing more than that.
Not a student management system
No enrolments, no certificates, no AVETMISS reporting. Apollo works alongside whatever you already run.
Not a plagiarism database
The integrity check compares the class in front of it and the sample answer. It cannot see the web or previous cohorts.
Not the decision-maker
Not a validation service
It gives your validators something explicit to review. It doesn’t do the validating.
How one trainer timed it, and what we haven’t proven.
Our marking-time figures come from one batch, timed by the trainer; estimates and extrapolations are labelled. Trainers’ own times are quoted as they gave them. Here’s the method, including where it falls short.
Not yet proven: one trainer, one batch. Marking quality hasn’t been independently checked.
Read it in your own numbers.
Run one unit through Apollo and measure your own turnaround and hours against last term.