JeddAI reads each student's work against the criteria, performance levels, and descriptors in your own rubric, then drafts feedback and scores that reference the specific evidence for each criterion. It also draws on your success criteria and comment banks, and you review and edit every comment before students see it.
What does JeddAI do when it reads my rubric?
When JeddAI reads your rubric, it treats each criterion, performance level, and descriptor as the instruction for what to look for in a student’s work. It then reads a response criterion by criterion, matches what the student wrote to the closest descriptor, and drafts a score plus a comment explaining that judgment. The rubric sets the standard; the tool applies it evenly across the class.
Because the descriptors define the judgment, the wording you use directly shapes the draft. A descriptor that names an observable skill, such as integrating evidence to support a claim, gives the tool something concrete to find. The result is feedback that points to a specific criterion rather than a broad impression, which is what makes a comment usable for the student who reads it.
What inputs shape the feedback besides the rubric?
Three inputs shape the feedback: your rubric, your success criteria, and your comment banks. The rubric defines the criteria and standards being graded; success criteria describe what a strong response looks like in student-friendly language; comment banks hold the phrases you reuse so the drafts sound like you. Together they give JeddAI a fuller picture than a rubric on its own.
You do not need all three from day one. The rubric does the heaviest lifting, and you can add success criteria and a short comment bank as you notice the guidance and phrasings you repeat. Each one narrows the gap between the draft and the feedback you would have written by hand, so later batches arrive closer to finished.
- Rubric: the criteria, performance levels, descriptors, and point values the work is graded against.
- Success criteria: a student-friendly description of what a strong response looks like.
- Comment banks: the feedback phrases you reuse so the drafts keep a consistent voice.
How does the feedback tie to specific parts of a student's work?
The feedback ties to specific parts of the work because JeddAI drafts it criterion by criterion, referencing the evidence in the response that led to each judgment. Instead of one overall grade with a vague note, you get a comment for each criterion that points to what the student did or did not do, mapped back to the descriptor it matched.
This evidence-tied approach is what separates rubric-aligned feedback from generic remarks. A comment noting that the thesis is stated but not sustained across paragraphs is actionable; one that says good effort is not. Because your rubric named sustaining a thesis as a criterion, the draft speaks to it directly, and the student knows exactly what to revise before the next attempt.
How does JeddAI keep feedback aligned to my standards and voice?
JeddAI keeps feedback aligned to your standards because you write the rubric, and it grades against your criteria exactly as worded. You can tie each criterion to the standard it assesses — a Common Core writing anchor standard, a state standard such as Texas TEKS or Florida B.E.S.T., or an AP scoring guideline — so the feedback stays standards-based and easy to justify.
Your voice comes from the comment banks. When you store the phrasings you already use, the drafts reach for those instead of a generic model tone, so students hear a consistent teacher across an assignment. Naming the standard inside the criterion also helps when a student, parent, or administrator asks how a judgment was reached: the rubric points straight back to the expectation you set.
What do I review before the feedback reaches a student?
You review every score and comment before anything reaches a student. JeddAI drafts; you read, adjust, and approve. Nothing is finalized automatically, and you can override any judgment, rewrite any comment, or change any score. The draft is a starting point that carries the routine work, not a decision made on your behalf.
In practice the first pass applies the rubric evenly and drafts comments that match your criteria, so your time goes to the calls that need a teacher: borderline responses, unusual answers, and the personal note a student will act on. You also decide how far to lean on the draft — a quick scan on low-stakes practice, a close read on a summative assessment.
How do I set things up so the drafts need less editing?
The drafts need less editing when your rubric is specific and your inputs reflect how you actually assess. Sharpen the descriptors so each performance level is clearly distinct and no two levels could describe the same paragraph. Test the rubric on two or three responses you have already graded; if the descriptors capture why one sits above another, they are precise enough to apply consistently.
Then let the setup learn your patterns. Reuse a rubric that works across similar assignments so grading stays even and you are not rebuilding it each unit. Watch where you edit the drafts most, because those edits usually point to a criterion or descriptor that was unclear, or a comment worth adding to your bank. Each refinement makes the next batch closer to what you would have written.
How does this give me consistent feedback and save time?
Consistency comes from the rubric applied the same way across a class, and the time saving comes from not drafting routine feedback from a blank page. When the same criteria and descriptors judge every response, students in every section are held to one standard, and your hours go to the comments that need your judgment rather than to repetitive phrasing.
A tool like JeddAI helps you apply your criteria evenly and cut the time spent drafting feedback, while you review and approve every score and comment. Built in Australia and used across AU, NZ, and internationally, it keeps the teacher in control. You can Get started with JeddAI and test your rubric on a single assignment before rolling it out to a class.
Frequently asked questions
Can JeddAI give feedback without a rubric?
It can start from success criteria or a comment bank, but the rubric is what makes the feedback consistent and standards-based. Adding one sharpens every draft and reduces how much you edit.
Does it draft written comments or just scores?
Both. It drafts a score and a written comment for each criterion, and the comment explains the score by pointing to the evidence in the student's work.
Can I use the same rubric across several sections?
Yes. Reusing one agreed rubric means every section is graded against the same criteria, which keeps feedback consistent and saves setup time each unit.
Will the drafted feedback sound generic?
Not if you use comment banks. Storing your own phrasings keeps drafts in your voice; without them the tone leans on a more general model style you would then rewrite.
How does it handle a holistic rubric rather than an analytic one?
Provide the holistic bands and their descriptors, and it drafts feedback against them, just as it would grade criterion by criterion with an analytic rubric.
Get started with Jeddle
Jeddle gives teachers and students instant, syllabus-aligned feedback powered by JeddAI.
Looking for study material? Browse Jeddle's Australian-English subject resources, or explore more articles on Using Jeddle.
