Poseidon brings AI-powered autograding, rubric-based review, and grade analytics into one platform, built for high schools, colleges, and universities grading everything from Python scripts to final papers.
Named for a good dog.
Student code runs against your test suite in a sandboxed environment. Scores, failure diagnostics, and output return automatically. Zero manual grading, any cohort size.
Design question-level rubrics with positive or negative scoring. Consistent, structured feedback on every submission. TAs work faster, students understand why.
Grade code, PDFs, essays, lab reports, and presentations in one platform. Every assignment type, every discipline. Not just computer science.
Item difficulty, discrimination indices, and quality flags on every rubric item. AI surfaces what your assessments are actually measuring, before you publish.
Release grades to a full cohort in one click. Score distributions, mean, median, and high surface automatically. Students get the full statistical picture, instantly.
Students file regrade requests directly in the platform, in context. Instructors review and respond with the original submission and grading visible side by side.
How it works
Step 1
Create courses, define rubrics, and configure your autograder. Invite students and TAs by email. They click a link and they are enrolled.
Step 2
Students upload submissions through the platform. The autograder runs in a sandboxed environment; TAs review and apply rubric items in the browser.
Step 3
Release grades to the full cohort in one click. Every student receives their score, class mean, median, and high directly in the platform.