SudoTask vs. Manually Grading Code Assignments
SudoTask is a classroom code-grading platform where teachers create programming tasks and students submit code that is graded automatically. It's for teachers who want to grade programming assignments consistently and give students fast feedback without running and checking every submission by hand. Here's how that compares to the manual approach.
| Aspect | Manually grading | SudoTask |
|---|---|---|
| Time per submission | You run each student's code yourself, test it against the examples, and compare the output by hand. | Each submission is run automatically against your test cases and the results show up on the submission page — you review results instead of running code. |
| Consistency of grading | Every submission is judged by eye, so the standard can drift between students, graders, or grading sessions. | Every submission is checked against the exact same test cases with the same criteria, so grading is uniform across the whole class. |
| Instant feedback for students | Students wait until you grade their assignment and return the results. | Students see pass or fail for each test case on the submission page once grading completes. |
| Scalability across large classes | Your workload grows with every extra submission you run and check. | Grading runs asynchronously in a queue, so one task definition grades any number of submissions; you only review the results page. |
| Hidden test cases | Keeping a separate set of hidden checks secret is awkward, and verifying them still means running code by hand. | Tasks support public and hidden test cases. Students see hidden cases only as pass or fail — never the expected output — so they can't reverse-engineer the grading. |
| Setup effort | None — you already have the assignment files. | One-time setup per task: write the prompt, add the input/output test cases, and SudoTask grades every submission from then on. |