Skip to main content
Assignment Evaluation suits work that comes in every day and has to be marked quickly. Here is how four kinds of customer use it.

​
Schools

The problem: Homework gets collected daily and marked days later, by which time the class has moved on.
Assignment Evaluation has no subject setting: everything is marked as English on the NCERT syllabus. It fits English work and anything you score against a rubric. For another subject where the marks matter, use Exam Evaluation.
How assignment evaluation helps:
  • Set up the worksheet once, upload the answer key, and then every student’s work is scored against the same answers and the same step marking.
  • Each answer comes back with the marks and feedback, not just a total, so the student knows what went wrong.
  • Students photograph their work on a phone, as PNG files. Every file you upload is checked, and a blurred one is refused with a message asking for a clearer picture.
  • No approval or publishing step to wait for. As soon as the AI finishes, the scores are readable.

​
Language and speaking programs

The problem: Spoken work is the slowest thing a teacher can mark, and the hardest to mark the same way twice. How assignment evaluation helps:
  • Students record their answer and you pass a link to the MP3 file. The AI scores the recording against a rubric.
  • Write the rubric once, with criteria such as fluency or vocabulary and a description of what earns each score band, and reuse it across assignments.
  • You get the words the AI heard along with the marks and an overall judgment for each criterion, so a teacher can see what the score is based on.
  • Build a criterion once and link it to several rubrics, instead of rewriting it per subject.

​
School software platforms

The problem: Your customers want homework marking inside the product they already use. How assignment evaluation helps:
  • Call the REST API from your backend and keep your own screens.
  • Tag every assignment and answer sheet with your own institute, course, batch, teacher, and student IDs, and list them back by any of those.
  • Register webhooks so your server is told when a worksheet has been read and when an answer sheet has been scored, instead of polling.
  • Fewer stages than exam evaluation, so less to build: there is no review step, no approval flag, and no publishing step.

​
Tutoring and practice apps

The problem: A student practicing alone gets no feedback until their next session. How assignment evaluation helps:
  • One assignment can hold as many answer sheets as you need, so the same worksheet serves every student who attempts it.
  • Rubric marking works for open-ended practice where there is no single right answer, such as a written explanation or a spoken summary.
  • The AI describes any diagram in a question, so diagram work can be set and marked.

​
Which product should I use?

​
Frequently asked questions

Not through this API. Assignment Evaluation has no endpoint for editing a score. If a teacher has to be able to change a mark, use Exam Evaluation.
No. Audio only, as an MP3 file at a public HTTPS address. Convert the recording to MP3 and host it somewhere reachable from the internet before you send the link.
No. A recording is treated as one student’s answer, so record one student per file.
No. You call REST endpoints and read JSON back. Your developers need to handle background work, which means either checking a status or receiving a webhook.
Yes. A rubric holds up to 20 criteria, and a criterion can be created on its own and linked to several rubrics. Change band wording carefully, because the AI reads it to work out what the criterion measures.

​
Next steps

Features

Rubrics, audio answers, and what comes back for each question.

How it works

The full flow from question set to scores.

Developer guide

Evaluate an assignment with the API, step by step.

Create a rubric

Build a rubric and its criteria with the API.